Selection of patients with hypogammaglobulinemia for immunoglobulin replacement therapy

Abstract

The present disclosure provides a method for selecting hypogammaglobulinemia patients in need of immunoglobulin replacement therapy (IgG-RT) by analyzing the patient's B cell repertoire. The method can be used before treating the patient with IgG-RT. The method can also include treating the patient. Also provided herein is a diagnostic product in a computer-readable medium that provides information for patient selection. The B cell repertoire can be measured by the number or abundance of individual antibody clones, by calculating a diversity index value for antibody clones, by calculating the total frequency of the 10 to 30 most frequent antibody clones, by determining the frequency of variable region gene (V region) usage in antibody clones, by measuring the percent germline identity of V or J regions in the BCR repertoire by comparing them with corresponding germline sequences, and / or by determining the somatic mutation frequency in different regions along the V region.

Description

[Technical Field]

[0001] 1. CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 63 / 439,054, filed January 13, 2023, which is incorporated herein by reference in its entirety.

[0002] 2. Sequence Listing This application contains a Sequence Listing that has been submitted electronically via the Patent Center and is incorporated herein by reference in its entirety. The XML copy, created on January 9, 2024, is named 54256WO_CRF_sequencelisting.xml and is 39,316 bytes in size. [Background technology]

[0003] 3. Background of the invention Defense against infection is orchestrated by a complex immune system in which all components have a task, and quantitative or qualitative defects in a single component often contribute to clinically evident immunodeficiency. The most common form of inborn error of immunity / primary immunodeficiency is antibody deficiency, a phenotype primarily characterized by recurrent upper respiratory tract infections. Antibody deficiencies include agammaglobulinemia (no antibodies), hypogammaglobulinemia (not enough antibodies), IgG subclass deficiency, and specific anti-PnPS (pneumococcal polysaccharide) deficiency, the latter of which manifests as recurrent pneumococcal infections.

[0004] Because immunoglobulin replacement preparations do not contain significant amounts of IgM or IgA, a combination of serum IgG levels and infection susceptibility was used to determine whether or not to provide IgG-RT. Therefore, IgG-RT is not indicated for the treatment of selective IgA deficiency. Although a decrease in IgG titer below 4 g / L is thought to be associated with an increased risk of infection, some patients with near-normal IgG levels may still exhibit pathological susceptibility to infection. Conversely, some people with IgG levels below 4 g / L do not exhibit significant susceptibility to infection, potentially because their immune system is able to respond to each challenge with high-quality acute naive and memory IgG responses.

[0005] Therefore, a clinical conundrum exists as to why some patients with severe hypogammaglobulinemia are not susceptible to infection and therefore do not require IgG-RT, while the majority of patients with antibody deficiencies require IgG-RT to remain healthy. There is a need to develop reliable methods for identifying patients who require immunoglobulin replacement therapy (IgG-RT). Summary of the Invention [Means for solving the problem]

[0006] 4. Summary of the Invention Applicant tested whether the composition of peripheral B cell receptor sequences and the number / diversity of B cell clones provide an indication of why some patients with severe hypogammaglobulinemia are not susceptible to infection and therefore do not require IgG-RT, while most patients with antibody deficiencies require IgG-RT to remain healthy. Specifically, Applicant sequenced and analyzed IgG and IgM heavy chain B cell receptor repertoires from PBMCs isolated from cohorts of patients with low serum IgG concentrations who either required IgG-RT or did not. Experimental data showed that patients requiring IgG-RT had more diverse IgG and IgM antibody repertoires, and their IgG sequences were significantly more similar to the germline. This suggests that although patients with low serum IgG concentrations who required IgG-RT had a more diverse repertoire, their antibody clones were less divergent from the germline and therefore may not be optimal for pathogen targeting, causing infection susceptibility. Conversely, those with low serum IgG concentrations that do not require IgG-RT have less diverse but also more mature antibody sequences that may be better suited to targeting pathogens.

[0007] Based on this work, the present application describes a method for selecting hypogammaglobulinemic patients for immune globulin replacement therapy (IgG-RT), comprising: (1) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, wherein each transcript encodes an IgG or IgM heavy chain or a portion thereof; (2) characterizing the patient's B cell receptor (BCR) repertoire by identifying antibody clones using the sequence information; (3)(a)~(f): (a) determining the number or abundance of individual antibody clones in said BCR repertoire; (b) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (c) selecting at least 10 but not more than 30 antibody clones that are most frequent in the BCR repertoire and calculating the total frequency of the 10 to 30 most frequent antibody clones; (d) determining variable region gene (V region) usage in the antibody clone, wherein optionally the V region gene is selected from the group consisting of an IGHV4-30-2 heavy chain V region gene, an IGHV4-30-4 heavy chain V region gene, an IGHV3-23 gene, an IGHV4-34 heavy chain V region gene, and an IGHV4-31 heavy chain V region gene; (e) determining percent germline identity by comparing the V or J regions in said BCR repertoire with corresponding germline sequences; and (f) determining the somatic mutation frequency at different regions along the V region; Analyzing the BCR repertoire by one or more selected from: (4)(g)~(s): (g) the number of IgG clones in the BCR repertoire is greater than Tg, wherein Tg is at least 600; (h) the diversity index value of the IgM antibody clones is greater than Th, wherein Th is at least 250; (i) the total frequency of the 10 to 30 most frequent IgG antibody clones in the BCR repertoire is less than Ti, where Ti is at most 30%; (j) the total frequency of the 10 to 30 most frequent IgM antibody clones in the BCR repertoire is less than Tj, wherein Tj is at most 7%; (k) the frequency of IgG antibody clones having the IGHV4-30-2 heavy chain V region is less than Tk, wherein Tk is at most 0.3%; (l) the frequency of IgG antibody clones having the IGHV4-30-4 heavy chain V region is less than Tl, wherein Tl is at most 0.5%; (m) the frequency of IgG antibody clones having the IGHV3-23 heavy chain V region is higher than Tm, wherein Tm is at least 6%; (n) the frequency of IgG antibody clones having the IGHV4-34 heavy chain V region is higher than Tn, wherein Tn is at least 6%; (o) the frequency of IgM antibody clones having the IGHV4-31 heavy chain V region is less than To, wherein To is at most 0.2%; (p) the average percent germline identity of the IgG V genes is greater than Tp, where Tp is at least 98%; (q) the average percent germline identity of the IgG J genes is greater than Tq, where Tq is at least 98%; (r) the median somatic nucleotide mutation frequency in the FR1, CDR1, FR2, or CDR2 region of IgG is less than Tr, where Tr is at most 1; and (s) the median somatic nucleotide mutation frequency in the FR3 region of IgG is less than Ts, where Ts is at most 4; selecting the patient for IgG-RT when one or more of the following conditions are met: (5) providing information regarding whether the patient is in need of IgG-RT; The present invention provides a method comprising:

[0008] In some embodiments, in step (1), sequence information for at least 50,000 transcripts is obtained. In some embodiments, in step (1), sequence information for at least 100,000, 500,000, 1,000,000, or 10 million transcripts is obtained.

[0009] In some embodiments, in step (4), the patient is selected for IgG-RT when (g) is satisfied. In some embodiments, in (g), Tg is 600, 650, 700, 750, 800, 850, 900, 950, 1,000, 1,050, or 1,100. In some embodiments, in (g), Tg is between 700 and 1,200, between 700 and 1,100, between 700 and 1,000, between 800 and 1,100, between 800 and 1,000, between 900 and 1,100, between 900 and 1,000, between 800 and 900, or between 700 and 800.

[0010] In some embodiments, in step (4), the patient is selected for IgG-RT when (h) is satisfied. In some embodiments, in (h), Th is 250, 300, 350, 400, 450, 500, 550, or 600. In some embodiments, in (h), Th is between 250 and 650, between 300 and 650, between 350 and 650, between 400 and 650, between 450 and 650, between 250 and 600, between 300 and 600, between 350 and 600, between 400 and 600, between 450 and 600, between 250 and 550, between 300 and 550, between 350 and 550, between 400 and 550, between 450 and 550, between 250 and 500, between 300 and 500, between 350 and 500, between 400 and 500, or between 450 and 500.

[0011] In some embodiments, in step (4), the patient is selected for IgG-RT when (i) is met. In some embodiments, in (i), Ti is 30%, 25%, or 20%. In some embodiments, in (i), Ti is between 20% and 35%, between 20% and 30%, between 20% and 25%, between 25% and 35%, or between 25% and 30%.

[0012] In some embodiments, in step (4), the patient is selected for IgG-RT when (j) is met. In some embodiments, in (j), Tj is 7%, 6.5%, 6%, 5.5%, 5%, 4.5%, or 4%. In some embodiments, in (j), Tj is between 4% and 7%, between 4% and 6%, between 4% and 5%, between 5% and 7%, between 5% and 6%, or between 6% and 7%.

[0013] In some embodiments, in step (4), the patient is selected for IgG-RT when (k) is satisfied. In some embodiments, in (k), Tk is 0.3%, 0.25%, 0.2%, or 0.15%. In some embodiments, in (k), Tk is between 0.15% and 0.3%, between 0.2% and 0.3%, between 0.25% and 0.3%, between 0.15% and 0.25%, between 0.15% and 0.2%, between 0.2% and 0.3%, or between 0.25% and 0.3%.

[0014] In some embodiments, in step (4), the patient is selected for IgG-RT when (l) is satisfied. In some embodiments, in (l), Tl is 0.5%, 0.45%, 0.4%, 0.35%, 0.3%, or 0.25%. In some embodiments, in (l), Tl is between 0.25% and 0.5%, between 0.25% and 0.45%, between 0.25% and 0.4%, between 0.25% and 0.35%, between 0.25% and 0.3%, between 0.3% and 0.5%, between 0.3% and 0.45%, between 0.3% and 0.4%, between 0.3% and 0.35%, between 0.35% and 0.45%, between 0.35% and 0.4%, between 0.4% and 0.5%, between 0.4% and 0.45%, or between 0.45% and 0.5%.

[0015] In some embodiments, in step (4), the patient is selected for IgG-RT when (m) is satisfied. In some embodiments, in (m), Tm is 6%, 7%, 8%, 9%, or 10%. In some embodiments, in (m), Tm is between 6% and 10%, between 6% and 9%, between 6% and 8%, between 6% and 7%, between 7% and 10%, between 7% and 8%, between 8% and 10%, between 8% and 9%, or between 9% and 10%.

[0016] In some embodiments, in step (4), the patient is selected for IgG-RT when (n) is met. In some embodiments, in (n), Tn is 6%, 7%, 8%, 9%, or 10%. In some embodiments, in (n), Tn is between 6% and 10%, between 6% and 9%, between 6% and 8%, between 6% and 7%, between 7% and 10%, between 7% and 8%, between 8% and 10%, between 8% and 9%, or between 9% and 10%.

[0017] In some embodiments, in step (4), the patient is selected for IgG-RT when (o) is met. In some embodiments, in (o), To is 0.2%, 0.15%, or 0.1%. In some embodiments, in (o), To is between 0.1% and 0.2%, between 0.1% and 0.15%, or between 0.15% and 0.2%.

[0018] In some embodiments, in step (4), the patient is selected for IgG-RT when (p) is met. In some embodiments, in (p), Tp is 98%, 98.5%, or 99%. In some embodiments, in (p), Tp is between 98% and 99%, between 98% and 98.5%, or between 98.5% and 99%.

[0019] In some embodiments, in step (4), the patient is selected for IgG-RT when (q) is met. In some embodiments, in (q), Tq is 98%, 98.5%, or 99%. In some embodiments, in (q), Tq is between 98% and 99%, between 98% and 98.5%, or between 98.5% and 99%.

[0020] In some embodiments, in step (4), the patient is selected for IgG-RT when (r) is satisfied. In some embodiments, in (r), Tr is 1, 0.8, or 0.6. In some embodiments, in (r), Tr is between 0.6 and 1, between 0.6 and 0.8, or between 0.8 and 1.

[0021] In some embodiments, in step (4), the patient is selected for IgG-RT when (s) is satisfied. In some embodiments, in (s), Ts is 4, 3.5, 3, 2.5, or 2. In some embodiments, in (s), Ts is between 2 and 4, between 2 and 3.5, between 2 and 3, between 2 and 2.5, between 2.5 and 4, between 2.5 and 3.5, between 2.5 and 3, between 3 and 4, between 3 and 3.5, or between 3.5 and 4.

[0022] In some embodiments, in step (3), one, two, three, four, five, or six analyses of (a) to (f) are performed for analyzing the BCR repertoire.

[0023] In some embodiments, in step (4), the patient is selected for IgG-RT when 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, or 13 criteria selected from (g) through (s) are met.

[0024] In some embodiments, in step (4), the patient is selected for IgG-RT when at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, or at least 12 criteria selected from (g) through (s) are met.

[0025] In some embodiments, the patient is selected for having less than 5 g / L of serum IgG and more than 40 / μL of peripheral B cells. In some embodiments, the patient is selected for having less than 4.5 g / L of serum IgG and more than 35 / μL of peripheral B cells. In some embodiments, the patient is selected for having less than 4 g / L of serum IgG and more than 30 / μL of peripheral B cells.

[0026] In some embodiments, the patient sample comprises peripheral blood mononuclear cells (PBMCs). In some embodiments, the method further comprises sequencing the at least 10,000 transcripts, thereby providing the sequence information. In some embodiments, the method further comprises, if the patient is selected for IgG-RT, treating the patient with IgG-RT.

[0027] In another aspect, the present disclosure provides a method of treating a patient with hypogammaglobulinemia, comprising administering immune globulin replacement therapy (IgG-RT) to the patient, wherein the patient has been selected for IgG-RT using the methods disclosed herein. In some embodiments, the method further comprises selecting the patient for IgG-RT using the methods disclosed herein.

[0028] In yet another aspect, the present disclosure provides a diagnostic product for selecting hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT), wherein said diagnostic product is stored on a non-transitory computer readable medium and comprises: (1) obtaining sequence information of a plurality of patients with hypogammaglobulinemia, wherein the sequence information of each patient comprises the sequences of at least 10,000 transcripts from a sample of said patient comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) characterizing the B cell receptor (BCR) repertoire of each of said patients by identifying antibody clones based on said sequence information; (3) obtaining a training dataset comprising a plurality of training examples, each training example corresponding to the BCR repertoire of an individual patient; and (a) one or more characteristics associated with the BCR repertoire of the individual patient; and (b) diagnosing said hypogammaglobulinemic patient as to whether said individual patient requires IgG-RT, optionally said diagnosis being based on information related to serum IgG levels or susceptibility to infection; and (4) numerically coding the training examples in the training dataset, including numerically coding the one or more characteristics associated with the BCR repertoire of the individual patient and numerically coding the diagnosis of the hypogammaglobulinemic patient as to whether the individual patient requires IgG-RT; (5) For a diagnostic model comprising a neural network having a plurality of layers, each layer having a plurality of parameters, the layer comprising an input layer for receiving the one or more numerically encoded features of the BCR repertoire, and an output layer indicative of the likelihood of need for IgG-RT, during one or more iterations of a training process: (a) applying the parameters of the neural network to a set of training examples for a current iteration to generate an estimated likelihood for the set of training examples; (b) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (c) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (d) stopping the backpropagation after the loss function meets a criterion; and (6) storing an updated set of parameters for the layer of the diagnostic model in the computer-readable storage medium; The present invention provides a diagnostic product manufactured by a process comprising:

[0029] In some embodiments, in step (3)(a), the one or more characteristics associated with the BCR repertoire of the individual patient are selected from 1) to 7), 1) the number or abundance of individual IgG clones in the BCR repertoire; 2) an IgM antibody clonal diversity index value calculated by measuring the number and abundance of individual IgM antibody clones within the BCR repertoire; 3) the total frequency of the top 10, top 11, top 12, top 13, top 14, top 15, top 16, top 17, top 18, top 19, top 20, top 21, top 22, top 23, top 24, top 25, top 26, top 27, top 28, top 29, or top 30 IgG antibody clones that are most frequently found in the BCR repertoire; 4) the total frequency of the top 10, top 11, top 12, top 13, top 14, top 15, top 16, top 17, top 18, top 19, top 20, top 21, top 22, top 23, top 24, top 25, top 26, top 27, top 28, top 29, or top 30 IgM antibody clones that are most frequently found in the BCR repertoire; 5) variable region (V) gene usage in the antibody clone, optionally wherein the V gene is selected from the group consisting of an IGHV4-30-2 heavy chain V gene, an IGHV4-30-4 heavy chain V gene, an IGHV3-23 heavy chain V gene, an IGHV4-34 heavy chain V gene, and an IGHV4-31 heavy chain V gene; 6) the average percent germline identity measured by comparing the variable (V) or joining (J) regions in the BCR repertoire to the corresponding germline sequences; and 7) the median somatic nucleotide mutation frequency in the V region in the BCR repertoire, optionally the V region comprising FR1, CDR1, FR2, CDR2 and / or FR3 regions; is selected from.

[0030] One aspect of the present disclosure is a diagnostic product for selecting hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT), wherein said diagnostic product is stored on a non-transitory computer readable medium and comprises: (1) obtaining sequence information of a plurality of patients with hypogammaglobulinemia, wherein the sequence information of each patient comprises the sequences of at least 10,000 transcripts from a sample of said patient comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) obtaining a training dataset comprising a plurality of training examples, wherein each training example comprises sequence information of said individual patient and a diagnosis of said individual patient as to whether said individual patient requires IgG-RT, optionally wherein said diagnosis is based on information related to serum IgG levels or susceptibility to infection; (3) numerically encoding the training examples in the training dataset, including numerically encoding the sequence information of the individual patient and numerically encoding the individual patient's diagnosis of whether the individual patient requires IgG-RT; (4) For a diagnostic model including a neural network having multiple layers, each layer having multiple parameters, the layers including an input layer for receiving the numerically coded sequence information and an output layer indicating the likelihood of need for IgG-RT, the diagnostic model is trained using the following training process: (a) applying parameters of the neural network to a set of training examples for a current iteration to generate an estimated likelihood for the set of training examples; (b) calculating a loss function indicating the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (c) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; (d) stopping the backpropagation after the loss function meets a criterion; For one or more iterations of (5) storing an updated set of parameters for the layer of the diagnostic model in the computer-readable storage medium; The present invention provides a diagnostic product manufactured by a process comprising:

[0031] Another aspect of the present disclosure is a method of selecting a hypogammaglobulinemic patient for immune globulin replacement therapy (IgG-RT), comprising: (1) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, wherein each of the transcripts encodes an IgG or IgM heavy chain or a portion thereof; (2) providing the sequence information or information related to the patient's B-cell receptor (BCR) repertoire obtained by processing the sequence information to a diagnostic product disclosed herein and operating the diagnostic product; (3) obtaining from the diagnostic product information regarding whether the patient requires IgG-RT; A method is disclosed, comprising:

[0032] In some embodiments, the step (2) of characterizing the B cell receptor (BCR) repertoire comprises (a) to (f): (a) determining the number or abundance of individual antibody clones in the BCR repertoire; (b) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (c) selecting at least 10 but not more than 30 antibody clones that are most frequent in the BCR repertoire and calculating the total frequency of the 10 to 30 frequent antibody clones; (d) determining variable region gene (V region) usage in said antibody clones, optionally wherein said V is selected from the group consisting of an IGHV4-30-2 heavy chain V region gene, an IGHV4-30-4 heavy chain V region gene, an IGHV3-23 heavy chain V region gene, an IGHV4-34 heavy chain V region gene, and an IGHV4-31 heavy chain V region gene; (e) determining percent germline identity by comparing the V or J regions in the BCR repertoire with corresponding germline sequences; and (f) determining somatic mutation frequencies at different regions along the V regions; The method comprises analyzing the BCR repertoire by one or more steps selected from:

[0033] In some embodiments, the method further comprises treating the patient with IgG-RT if the patient is selected for IgG-RT.

[0034] The present disclosure also provides a method of treating a patient with hypogammaglobulinemia, comprising administering immune globulin replacement therapy (IgG-RT) to the patient, wherein the patient has been selected for IgG-RT using the methods disclosed herein. [Brief explanation of the drawings]

[0035] 5. A brief description of some figures in the drawing These and other features, aspects, and advantages of the present invention will become better understood with regard to the following description and accompanying drawings.

[0036] [Figure 1A-B] Figure 1A-1F. IgG and IgM antibody repertoire sequencing. (Figure 1A) IgG (left panel), IgM (center panel), and IgA (right panel) antibody titers for patients who required IgG-RT and patients who did not. (Figure 1B) Number of IgG and IgM antibody clones for patients who required IgG-RT and patients who did not. (Figure 1C) IgG and IgM antibody diversity index for patients who required IgG-RT and patients who did not. (Figure 1D) Cumulative frequency of the top 20 IgG clones (each patient is a different color). The y-axis shows cumulative frequency measured as a percent of the total repertoire, and the x-axis shows the top 20 clones ordered from most abundant to least abundant. The right and left panels show patients who required IgG-RT and patients who did not require IgG-RT, respectively. (Figure 1E) Cumulative frequency of the top 20 IgM clones (each patient is a different color). (FIG. 1F) Heavy chain CDR3 amino acid length distribution of IgG (left panel) and IgM (right panel). [Figure 1C] Figure 1A-1F. IgG and IgM antibody repertoire sequencing. (Figure 1A) IgG (left panel), IgM (center panel), and IgA (right panel) antibody titers for patients who required IgG-RT and patients who did not. (Figure 1B) Number of IgG and IgM antibody clones for patients who required IgG-RT and patients who did not. (Figure 1C) IgG and IgM antibody diversity index for patients who required IgG-RT and patients who did not. (Figure 1D) Cumulative frequency of the top 20 IgG clones (each patient is a different color). The y-axis shows cumulative frequency measured as a percent of the total repertoire, and the x-axis shows the top 20 clones ordered from most abundant to least abundant. The right and left panels show patients who required IgG-RT and patients who did not require IgG-RT, respectively. (Figure 1E) Cumulative frequency of the top 20 IgM clones (each patient is a different color). (FIG. 1F) Heavy chain CDR3 amino acid length distribution of IgG (left panel) and IgM (right panel). [Figure 1D-F]Figure 1A-1F. IgG and IgM antibody repertoire sequencing. (Figure 1A) IgG (left panel), IgM (center panel), and IgA (right panel) antibody titers for patients who required IgG-RT and patients who did not. (Figure 1B) Number of IgG and IgM antibody clones for patients who required IgG-RT and patients who did not. (Figure 1C) IgG and IgM antibody diversity index for patients who required IgG-RT and patients who did not. (Figure 1D) Cumulative frequency of the top 20 IgG clones (each patient is a different color). The y-axis shows cumulative frequency measured as a percent of the total repertoire, and the x-axis shows the top 20 clones ordered from most abundant to least abundant. The right and left panels show patients who required IgG-RT and patients who did not require IgG-RT, respectively. (Figure 1E) Cumulative frequency of the top 20 IgM clones (each patient is a different color). (FIG. 1F) Heavy chain CDR3 amino acid length distribution of IgG (left panel) and IgM (right panel).

[0037] [Figure 2A] Figures 2A-2J. Correlations between antibody repertoires and immune features. (Figures 2A and 2B) Inter-element correlation matrices for various antibody features and immune cell frequencies for patients who did not require (Figure 2A) and did require (Figure 2B) IgG-RT. Numbers indicate Pearson correlation coefficients. Blue and red shading indicates positive and negative correlations, respectively, as indicated in the legend of Figure 2B. Only significant (p ≤ 0.05) correlations are shown. (Figures 2C-2J) Scatterplots showing several significant correlations from Figures 2A and 2B for patients who did require (right panel) and did not require (left panel) IgG-RT. The blue line is the linear regression line, and gray shading indicates the 95% confidence interval around the fitted line. P values ​​are indicated in black (p > 0.05) or red (p ≤ 0.05). [Figure 2B]Figures 2A-2J. Correlations between antibody repertoires and immune features. (Figures 2A and 2B) Inter-element correlation matrices for various antibody features and immune cell frequencies for patients who did not require (Figure 2A) and did require (Figure 2B) IgG-RT. Numbers indicate Pearson correlation coefficients. Blue and red shading indicates positive and negative correlations, respectively, as indicated in the legend of Figure 2B. Only significant (p ≤ 0.05) correlations are shown. (Figures 2C-2J) Scatterplots showing several significant correlations from Figures 2A and 2B for patients who did require (right panel) and did not require (left panel) IgG-RT. The blue line is the linear regression line, and gray shading indicates the 95% confidence interval around the fitted line. P values ​​are indicated in black (p > 0.05) or red (p ≤ 0.05). [Figure 2C-F] Figures 2A-2J. Correlations between antibody repertoires and immune features. (Figures 2A and 2B) Inter-element correlation matrices for various antibody features and immune cell frequencies for patients who did not require (Figure 2A) and did require (Figure 2B) IgG-RT. Numbers indicate Pearson correlation coefficients. Blue and red shading indicates positive and negative correlations, respectively, as indicated in the legend of Figure 2B. Only significant (p ≤ 0.05) correlations are shown. (Figures 2C-2J) Scatterplots showing several significant correlations from Figures 2A and 2B for patients who did require (right panel) and did not require (left panel) IgG-RT. The blue line is the linear regression line, and gray shading indicates the 95% confidence interval around the fitted line. P values ​​are indicated in black (p > 0.05) or red (p ≤ 0.05). [Figure 2G-J]Figures 2A-2J. Correlations between antibody repertoires and immune features. (Figures 2A and 2B) Inter-element correlation matrices for various antibody features and immune cell frequencies for patients who did not require (Figure 2A) and did require (Figure 2B) IgG-RT. Numbers indicate Pearson correlation coefficients. Blue and red shading indicates positive and negative correlations, respectively, as indicated in the legend of Figure 2B. Only significant (p ≤ 0.05) correlations are shown. (Figures 2C-2J) Scatterplots showing several significant correlations from Figures 2A and 2B for patients who did require (right panel) and did not require (left panel) IgG-RT. The blue line is the linear regression line, and gray shading indicates the 95% confidence interval around the fitted line. P values ​​are indicated in black (p > 0.05) or red (p ≤ 0.05).

[0038] [Figure 3A-B]Figures 3A-3G. Diversity of antibody heavy chain V and J genes. (Figure 3A) Heatmap showing the abundance (y-axis) of antibody clones with specific heavy chain V genes for patients (x-axis) who required (right panel) and did not require (left panel) IgG-RT for IgG. Color indicates clonal frequency per patient, as indicated by the legend. (Figure 3B) Heatmap showing IgM heavy chain V gene usage. (Figure 3C) IgG heavy chain V genes present at different frequencies between patients who required and did not require IgG-RT. The y-axis represents the percentage of antibody clones with a given V gene. P values ​​are adjusted using the Benjamini-Hochberg method to correct for multiple testing. (Figure 3D) Differences in IgM heavy chain V gene usage between patients who required and did not require IgG-RT. (Figure 3E) Boxplot showing the percent nucleotide identity of V and J genes relative to germline sequences for all IgG clones. (Figure 3F) Boxplot showing the percent amino acid identity of V and J genes relative to germline sequences for all IgM clones. (Figure 3G) V gene nucleotide mutation frequencies in different regions for IgG (left panel) and IgM (right panel). FR = framework; CDR = complementarity-determining region. ns (not significant): p>0.05, *: p≦0.05, **: p≦0.01, ***: p≦0.001, ****: p≦0.0001. [Figure 3C-G]Figures 3A-3G. Diversity of antibody heavy chain V and J genes. (Figure 3A) Heatmap showing the abundance (y-axis) of antibody clones with specific heavy chain V genes for patients (x-axis) who required (right panel) and did not require (left panel) IgG-RT for IgG. Color indicates clonal frequency per patient, as indicated by the legend. (Figure 3B) Heatmap showing IgM heavy chain V gene usage. (Figure 3C) IgG heavy chain V genes present at different frequencies between patients who required and did not require IgG-RT. The y-axis represents the percentage of antibody clones with a given V gene. P values ​​are adjusted using the Benjamini-Hochberg method to correct for multiple testing. (Figure 3D) Differences in IgM heavy chain V gene usage between patients who required and did not require IgG-RT. (Figure 3E) Boxplot showing the percent nucleotide identity of V and J genes relative to germline sequences for all IgG clones. (Figure 3F) Boxplot showing the percent amino acid identity of V and J genes relative to germline sequences for all IgM clones. (Figure 3G) V gene nucleotide mutation frequencies in different regions for IgG (left panel) and IgM (right panel). FR = framework; CDR = complementarity-determining region. ns (not significant): p>0.05, *: p≦0.05, **: p≦0.01, ***: p≦0.001, ****: p≦0.0001.

[0039] [Figure 4A-C] Figures 4A-4C. IgG and IgM antibody repertoire sequencing. The analyses in Figures 1A-1C were repeated after removing two higher titer donors (IgG titers >4 g / L or IgM titers >1 g / L). (Figure 4A) IgG, IgM, and IgA antibody titers for patients who required IgG-RT and those who did not. (Figure 4B) Number of IgG and IgM antibody clones. (Figure 4C) IgG and IgM antibody diversity index.

[0040] [Figure 5-1]Figure 5. Scatterplots showing several significant correlations between various antibody characteristics and immune cell frequencies for patients who required IgG-RT (right panel) and those who did not (left panel). The blue line is the linear regression line, and the gray shading indicates the 95% confidence interval around the fitted line. P values ​​are shown in black (p>0.05) or red (p≦0.05). [Figure 5-2] Figure 5. Scatterplots showing several significant correlations between various antibody characteristics and immune cell frequencies for patients who required IgG-RT (right panel) and those who did not (left panel). The blue line is the linear regression line, and the gray shading indicates the 95% confidence interval around the fitted line. P values ​​are shown in black (p>0.05) or red (p≦0.05).

[0041] [Figure 6A-B] Figures 6A-6D. (Figure 6A) Heatmap showing the abundance of antibody clones with specific heavy chain J genes (y-axis) for patients (x-axis) who required IgG-RT for IgG (right panel) and did not require IgG-RT for IgG (left panel). Color indicates clonal frequency per patient, as indicated by the legend. (Figure 6B) Heatmap showing IgM heavy chain J gene usage. (Figure 6C) Principal component analysis (PCA) using IgG heavy chain V gene usage. Data points represent individual patients colored based on their need for IgG-RT, as indicated in the legend. (Figure 6D) PCA using IgM V gene usage. [Figure 6C-D] Figures 6A-6D. (Figure 6A) Heatmap showing the abundance of antibody clones with specific heavy chain J genes (y-axis) for patients (x-axis) who required IgG-RT for IgG (right panel) and did not require IgG-RT for IgG (left panel). Color indicates clonal frequency per patient, as indicated by the legend. (Figure 6B) Heatmap showing IgM heavy chain J gene usage. (Figure 6C) Principal component analysis (PCA) using IgG heavy chain V gene usage. Data points represent individual patients colored based on their need for IgG-RT, as indicated in the legend. (Figure 6D) PCA using IgM V gene usage.

[0042] [Figure 7A-C] Figures 7A-7E. Divergence of antibody heavy chain V and J genes from germline. (Figure 7A) Boxplots showing percent nucleotide identity of V and J genes to germline sequences for IgG. Replicate analysis of Figure 3E after removing two high-titer donors. (Figure 7B) Boxplots showing percent amino acid identity of V and J genes to germline sequences for IgM. Replicate analysis of Figure 3F after removing two high-titer donors. (Figure 7C) V gene nucleotide mutation frequencies in different regions for IgG (left panel) and IgM (right panel). Replicate analysis of Figure 3G after removing two high-titer donors. (Figure 7D) V gene amino acid mutation frequencies in different regions for IgG (left panel) and IgM (right panel). (Figure 7E) Replicate analysis of 7D with removal of two high-titer donors. FR = framework; CDR = complementarity-determining region. ns (not significant): p>0.05, *: p≦0.05, **: p≦0.01, ***: p≦0.001, ****: p≦0.0001. [Figure 7D-E] Figures 7A-7E. Divergence of antibody heavy chain V and J genes from germline. (Figure 7A) Boxplots showing percent nucleotide identity of V and J genes to germline sequences for IgG. Replicate analysis of Figure 3E after removing two high-titer donors. (Figure 7B) Boxplots showing percent amino acid identity of V and J genes to germline sequences for IgM. Replicate analysis of Figure 3F after removing two high-titer donors. (Figure 7C) V gene nucleotide mutation frequencies in different regions for IgG (left panel) and IgM (right panel). Replicate analysis of Figure 3G after removing two high-titer donors. (Figure 7D) V gene amino acid mutation frequencies in different regions for IgG (left panel) and IgM (right panel). (Figure 7E) Replicate analysis of 7D with removal of two high-titer donors. FR = framework; CDR = complementarity-determining region. ns (not significant): p>0.05, *: p≦0.05, **: p≦0.01, ***: p≦0.001, ****: p≦0.0001.

[0043] [Figure 8-1]Figure 8. V gene nucleotide mutation frequencies along the top five most common heavy chain V genes for IgG (top panel) and IgM (bottom panel), for patients who required IgG-RT (right panel) and those who did not (left panel). The x-axis indicates the nucleotide position along the V gene, and the y-axis indicates the nucleotide mutation frequency (number of nucleotide mutations per sequencing read). Individual donors are color-coded as indicated in the legend. Background shading indicates framework (FR) and complementarity-determining regions (CDR). The first 21 nucleotides (primer binding sites) were excluded from the analysis. [Figure 8-2] Figure 8. V gene nucleotide mutation frequencies along the top five most common heavy chain V genes for IgG (top panel) and IgM (bottom panel), for patients who required IgG-RT (right panel) and those who did not (left panel). The x-axis indicates the nucleotide position along the V gene, and the y-axis indicates the nucleotide mutation frequency (number of nucleotide mutations per sequencing read). Individual donors are color-coded as indicated in the legend. Background shading indicates framework (FR) and complementarity-determining regions (CDR). The first 21 nucleotides (primer binding sites) were excluded from the analysis.

[0044] [Figure 9-1] Figure 9. V gene amino acid mutation frequencies along the top five most common heavy chain V genes for IgG (top panel) and IgM (bottom panel), for patients who required IgG-RT (right panel) and did not require IgG-RT (right panel). The x-axis indicates the amino acid position along the V gene, and the y-axis indicates the amino acid mutation frequency (number of amino acid mutations per sequencing read). Individual donors are color-coded as indicated in the legend. Background shading indicates framework (FR) and complementarity-determining regions (CDR). The first seven amino acids (primer binding site) were excluded from the analysis. [Figure 9-2]Figure 9. V gene amino acid mutation frequencies along the top five most common heavy chain V genes for IgG (top panel) and IgM (bottom panel), for patients who required IgG-RT (right panel) and did not require IgG-RT (right panel). The x-axis indicates the amino acid position along the V gene, and the y-axis indicates the amino acid mutation frequency (number of amino acid mutations per sequencing read). Individual donors are color-coded as indicated in the legend. Background shading indicates framework (FR) and complementarity-determining regions (CDR). The first seven amino acids (primer binding site) were excluded from the analysis.

[0045] [Figure 10A] Figures 10A-10B. IgG IGHV4-34 Mutations. (Figure 10A) Amino acid mutation frequencies along IgG IGHV4-34 for patients who required IgG-RT (lower panel) and patients who did not require IgG-RT (upper panel). The x-axis indicates the amino acid position along the V gene, and the y-axis indicates the amino acid mutation frequency (number of amino acid mutations per sequencing read). Individual donors are color-coded as indicated in the legend. Background shading indicates framework (FR) and complementarity-determining regions (CDR). The first 7 amino acids (primer binding site) were excluded from the analysis. The protein sequence is shown, with the hydrophobic patch (AVY residues) in red. (Figure 10B) Average mutation frequency in the IGHV4-34 hydrophobic patch (AVY residues). Figure 10A discloses SEQ ID NOs: 18 and 18, respectively, in order of appearance. [Figure 10B]Figures 10A-10B. IgG IGHV4-34 Mutations. (Figure 10A) Amino acid mutation frequencies along IgG IGHV4-34 for patients who required IgG-RT (lower panel) and patients who did not require IgG-RT (upper panel). The x-axis indicates the amino acid position along the V gene, and the y-axis indicates the amino acid mutation frequency (number of amino acid mutations per sequencing read). Individual donors are color-coded as indicated in the legend. Background shading indicates framework (FR) and complementarity-determining regions (CDR). The first 7 amino acids (primer binding site) were excluded from the analysis. The protein sequence is shown, with the hydrophobic patch (AVY residues) in red. (Figure 10B) Average mutation frequency in the IGHV4-34 hydrophobic patch (AVY residues). Figure 10A discloses SEQ ID NOs: 18 and 18, respectively, in order of appearance. DETAILED DESCRIPTION OF THE INVENTION

[0046] 6. Detailed Description of the Invention 6.2 How to Select Patients with Hypogammaglobulinemia for Immunoglobulin Replacement Therapy (Ig-RT) The present disclosure relates to a method for selecting a hypogammaglobulinemic patient for immunoglobulin replacement therapy (IgG-RT). In some embodiments, the method is used prior to immunoglobulin replacement therapy (IgG-RT). The method can use sequence information associated with the patient's B cells. Thus, the method can further include obtaining the sequence information. In some embodiments, the method includes sequencing at least 10,000 transcripts from the patient, thereby providing the sequence information. In some embodiments, the sequence information relates to at least 10,000 transcripts from a patient sample comprising B cells. In some embodiments, the patient sample comprises peripheral blood mononuclear cells (PBMCs).

[0047] In some embodiments, the method comprises: (2) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, each transcript encoding an IgG or IgM heavy chain or a portion thereof; (3) characterizing the patient's B cell receptor (BCR) repertoire by identifying antibody clones using the sequence information; (4) analyzing the BCR repertoire; (5) selecting patients for IgG-RT based on analysis of the BCR repertoire; (6) providing information regarding whether the patient requires IgG-RT; Includes:

[0048] In some embodiments, the method comprises: (1) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, wherein each transcript encodes an IgG or IgM heavy chain or a portion thereof; (2) characterizing the patient's B cell receptor (BCR) repertoire by identifying antibody clones using the sequence information; (3)(a)~(f): (a) determining the number or abundance of individual antibody clones in said BCR repertoire; (b) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (c) selecting at least 10 but not more than 30 antibody clones that are most frequent in the BCR repertoire and calculating the total frequency of the 10 to 30 most frequent antibody clones; (d) determining variable region gene (V region) usage in the antibody clone, wherein optionally the V region gene is selected from the group consisting of an IGHV4-30-2 heavy chain V region gene, an IGHV4-30-4 heavy chain V region gene, an IGHV3-23 gene, an IGHV4-34 heavy chain V region gene, and an IGHV4-31 heavy chain V region gene; (e) determining percent germline identity by comparing the V or J regions in said BCR repertoire with corresponding germline sequences; and (f) determining the somatic mutation frequency at different regions along the V region; Analyzing the BCR repertoire by one or more selected from: (4)(g)~(s): (g) the number of IgG clones in the BCR repertoire is greater than Tg, wherein Tg is at least 600; (h) the diversity index value of the IgM antibody clones is greater than Th, wherein Th is at least 250; (i) the total frequency of the 10 to 30 most frequent IgG antibody clones in the BCR repertoire is less than Ti, where Ti is at most 30%; (j) the total frequency of the 10 to 30 most frequent IgM antibody clones in the BCR repertoire is less than Tj, wherein Tj is at most 7%; (k) the frequency of IgG antibody clones having the IGHV4-30-2 heavy chain V region is less than Tk, wherein Tk is at most 0.3%; (l) the frequency of IgG antibody clones having the IGHV4-30-4 heavy chain V region is less than Tl, wherein Tl is at most 0.5%; (m) the frequency of IgG antibody clones having the IGHV3-23 heavy chain V region is higher than Tm, wherein Tm is at least 6%; (n) the frequency of IgG antibody clones having the IGHV4-34 heavy chain V region is higher than Tn, wherein Tn is at least 6%; (o) the frequency of IgM antibody clones having the IGHV4-31 heavy chain V region is less than To, wherein To is at most 0.2%; (p) the average percent germline identity of the IgG V genes is greater than Tp, where Tp is at least 98%; (q) the average percent germline identity of the IgG J genes is greater than Tq, where Tq is at least 98%; (r) the median somatic nucleotide mutation frequency in the FR1, CDR1, FR2, or CDR2 region of IgG is less than Tr, where Tr is at most 1; and (s) the median somatic nucleotide mutation frequency in the FR3 region of IgG is less than Ts, where Ts is at most 4; selecting the patient for IgG-RT when one or more of the following conditions are met: (5) providing information regarding whether the patient is in need of IgG-RT; Includes:

[0049] 6.5.1 Sequence information from patient B-cell transcripts In some embodiments, the methods disclosed herein use sequence information of at least 10,000 transcripts from a patient sample containing B cells.

[0050] In some embodiments, the sequence information is obtained by sequencing a sample from the patient. In some embodiments, the sequence information is obtained from a database.

[0051] In some embodiments, the sequence information is sequence information for at least 500 transcripts. In some embodiments, the sequence information is sequence information for at least 1000 transcripts. In some embodiments, the sequence information is sequence information for at least 5000 transcripts. In some embodiments, the sequence information is sequence information for at least 10,000 transcripts. In some embodiments, the sequence information is sequence information for at least 50,000 transcripts. In some embodiments, the sequence information is sequence information for at least 100,000, 500,000, 1,000,000, or 10 million transcripts. In some embodiments, the sequence information is sequence information for more than 10 million transcripts.

[0052] In some embodiments, each transcript encodes an IgG or IgM heavy chain or a portion thereof. In some embodiments, each transcript encodes an IgG heavy chain or a portion thereof. In some embodiments, each transcript encodes an IgM heavy chain or a portion thereof. In some embodiments, the transcripts as a group encode IgG and IgM heavy chains. In some embodiments, each transcript encodes the CDR3 of an IgG heavy chain. In some embodiments, each transcript encodes the CDR3 of an IgM heavy chain. In some embodiments, each transcript encodes the CDR3 of an IgG or IgM heavy chain.

[0053] 6.5.2 Identification of antibody clones In various embodiments, an antibody clone, as used herein, refers to a homogeneous antibody derived from a single B cell that detects a single epitope within an immunogen. In some embodiments, a clone is defined conservatively, and an antibody clone can include antibodies with one or one to two amino acid differences in the CDR3 region. In some embodiments, unique sequences are combined into a clone if they have one amino acid difference in the CDR3H. In some embodiments, unique sequences are combined if they have one to two amino acid differences in the CDR3H. In some embodiments, unique sequences are combined if they have one amino acid difference in a CDR3H that is 5 to 6 amino acids long, or if they have one to two amino acid differences in a CDR3H that is longer than 6 amino acids.

[0054] Antibody clones can be identified by a variety of methods known in the art, such as single cell techniques (some involving microfluidic technology).

[0055] In some embodiments, sequence information can be used to identify antibody clones. Specifically, antibody clones can be identified by analyzing sequence information of transcripts from patient samples containing B cells. In some embodiments, sequence information related to IgG or IgM heavy chains or portions thereof is analyzed. In some embodiments, sequences corresponding to CDR3 of IgG or IgM heavy chains are used to identify antibody clones. In some embodiments, sequences corresponding to the variable regions of IgG or IgM heavy or light chains are used to identify antibody clones.

[0056] 6.5.3 Analysis of the B-cell repertoire The method involves analysis of the B cell repertoire in the patient.

[0057] In some embodiments, the number or abundance of individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 5 individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 10 individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 20 individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 50 individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 100 individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 500 individual antibody clones in the BCR repertoire is measured. In some embodiments, the number or abundance of at least 1000 individual antibody clones in the BCR repertoire is measured.

[0058] In some embodiments, the diversity index value of an antibody clone is calculated by measuring the number and abundance of individual antibody clones in a BCR repertoire. The diversity index value can be calculated by the method described in Example 6.1 ("Antibody Diversity Index"). Specifically, the antibody diversity index can be calculated using the diversity function of the tcR package (version 2.3.2) in R version 4.1.2. The true diversity of an antibody repertoire X refers to the effective richness of that population, i.e., the number of equally common antibody clones that would be required to produce a repertoire with the same overall diversity as X. This value increases with the number of antibody clones in the repertoire and with the uniformity with which these clones are distributed.

[0059] In some embodiments, the method comprises selecting at least 10 but not more than 30 of the most frequent antibody clones in the BCR repertoire and calculating a total frequency of the 10 to 30 most frequent antibody clones. In some embodiments, the 10 most frequent antibody clones are selected. In some embodiments, 15 most frequent antibody clones are selected. In some embodiments, 20 most frequent antibody clones are selected. In some embodiments, 25 most frequent antibody clones are selected. In some embodiments, 30 most frequent antibody clones are selected. In some embodiments, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 most frequent antibody clones are selected. In some embodiments, 10 to 15 most frequent antibody clones are selected. In some embodiments, 15 to 20 most frequent antibody clones are selected. In some embodiments, the 20 to 25 most frequent antibody clones are selected, hi some embodiments, the 25 to 30 most frequent antibody clones are selected.

[0060] In some embodiments, a method comprises determining the frequency of variable region gene (V region) usage in an antibody clone. In some embodiments, the method comprises determining the frequency of V region gene usage selected from the group consisting of IGHV4-30-2 heavy chain V region genes, IGHV4-30-4 heavy chain V region genes, IGHV3-23 genes, IGHV4-34 heavy chain V region genes, and IGHV4-31 heavy chain V region genes. In some embodiments, the method comprises determining the frequency of IGHV4-30-2 heavy chain V region gene usage. In some embodiments, the method comprises determining the frequency of IGHV4-30-4 heavy chain V region gene usage. In some embodiments, the method comprises determining the frequency of IGHV3-23 gene usage. In some embodiments, the method comprises determining the frequency of IGHV4-34 heavy chain V region gene usage. In some embodiments, the method comprises determining the frequency of IGHV4-31 heavy chain V region gene usage.

[0061] In some embodiments, the method comprises measuring percent germline identity by comparing the V or J region in the BCR repertoire with the corresponding germline sequence. The identity of germline V and J genes can be measured by mapping the antibody nucleotide sequence to human V and J gene reference sequences. In some embodiments, UBLAST alignment is used to assign V and J gene families and calculate percent identity to the germline sequence.

[0062] In some embodiments, the methods include determining somatic mutation frequencies at different regions along the V region.

[0063] In some embodiments, the method comprises: (a) through (f): (a) determining the number or abundance of individual antibody clones in the BCR repertoire; (b) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (c) selecting at least 10 but not more than 30 antibody clones that are most frequent in the BCR repertoire and calculating the total frequency of the 10 to 30 most frequent antibody clones; (d) determining variable region gene (V region) usage in said antibody clones, optionally wherein said V region genes are selected from the group consisting of IGHV4-30-2 heavy chain V region genes, IGHV4-30-4 heavy chain V region genes, IGHV3-23 genes, IGHV4-34 heavy chain V region genes, and IGHV4-31 heavy chain V region genes; (e) determining percent germline identity by comparing the V or J regions in the BCR repertoire with corresponding germline sequences; and (f) determining somatic mutation frequencies at different regions along the V regions; including one or more selected from;

[0064] In some embodiments, one of (a)-(f) is performed for the analysis of the BCR repertoire. In some embodiments, two of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, three of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, four of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, five of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, six of (a)-(f) are performed for the analysis of the BCR repertoire.

[0065] In some embodiments, at least one of (a)-(f) is performed for the analysis of the BCR repertoire. In some embodiments, at least two of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, at least three of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, at least four of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, at least five of (a)-(f) are performed for the analysis of the BCR repertoire. In some embodiments, all six of (a)-(f) are performed for the analysis of the BCR repertoire.

[0066] 6.5.4 Patient Selection for IgG-RT In various embodiments, the method includes selecting the patient for IgG-RT based on analysis of the B cell repertoire, hi some embodiments, the patient is selected for IgG-RT if the patient has a B cell repertoire similar to one or more patients clinically documented to be in need of IgG-RT.

[0067] In some embodiments, patients are selected for IgG-RT when the number of IgG clones in their BCR repertoire is greater than a threshold g (Tg), where Tg is at least 600. In some embodiments, Tg is 600, 650, 700, 750, 800, 850, 900, 950, 1,000, 1,050, or 1,100. In some embodiments, patients are selected for IgG-RT when the number of IgG clones in their BCR repertoire is greater than a threshold g (Tg), where Tg is at least 600, 650, 700, 750, 800, 850, 900, 950, 1,000, 1,050, or 1,100. In some embodiments, patients are selected for IgG-RT if the number of IgG clones in the BCR repertoire is greater than a threshold g (Tg), where Tg is a number between 600 and 2000, between 800 and 1500, between 1000 and 1500, or between 1100 and 1300. In some embodiments, Tg is a number between 700 and 1,200, between 700 and 1,100, between 700 and 1,000, between 800 and 1,100, between 800 and 1,000, between 900 and 1,100, between 900 and 1,000, between 800 and 900, or between 700 and 800.

[0068] In some embodiments, patients are selected for IgG-RT if the diversity index value of IgM antibody clones is greater than a threshold h(Th), where Th is at least 250. In some embodiments, Th is 250, 300, 350, 400, 450, 500, 550, or 600. In some embodiments, Th is 250, 300, 350, 400, 450, 500, 550, or 600. In some embodiments, Th is a number between 250 and 650, between 300 and 650, between 350 and 650, between 400 and 650, between 450 and 650, between 250 and 600, between 300 and 600, between 350 and 600, between 400 and 600, between 450 and 600, between 250 and 550, between 300 and 550, between 350 and 550, between 400 and 550, between 450 and 550, between 250 and 500, between 300 and 500, between 350 and 500, between 400 and 500, or between 450 and 500.

[0069] In some embodiments, patients are selected for IgG-RT, and the total frequency of the 10 to 30 most frequent IgG antibody clones in the BCR repertoire is less than a threshold i(Ti), where Ti is at most 30%. In some embodiments, Ti is 30%, 25%, or 20%. In some embodiments, Ti is between 20% and 35%, between 20% and 30%, between 20% and 25%, or between 25% and 35%, or between 25% and 30%.

[0070] In some embodiments, patients are selected for IgG-RT when the combined frequency of the 10 to 30 most frequent IgM antibody clones in the BCR repertoire is below a threshold j (Tj), where Tj is at most 7%. In some embodiments, Tj is 7%, 6.5%, 6%, 5.5%, 5%, 4.5%, or 4%. In some embodiments, Tj is between 4% and 7%, between 4% and 6%, between 4% and 5%, between 5% and 7%, between 5% and 6%, or between 6% and 7%.

[0071] In some embodiments, patients are selected for IgG-RT if the frequency of IgG antibody clones having an IGHV4-30-2 heavy chain V region is below a threshold k (Tk), where Tk is at most 0.3%. In some embodiments, Tk is 0.3%, 0.25%, 0.2%, or 0.15%. In some embodiments, Tk is between 0.15% and 0.3%, between 0.2% and 0.3%, between 0.25% and 0.3%, between 0.15% and 0.25%, between 0.15% and 0.2%, between 0.2% and 0.3%, or between 0.25% and 0.3%.

[0072] In some embodiments, patients are selected for IgG-RT if the frequency of IgG antibody clones having an IGHV4-30-4 heavy chain V region is below a threshold value l (Tl), where Tl is at most 0.5%. In some embodiments, Tl is 0.5%, 0.45%, 0.4%, 0.35%, 0.3%, or 0.25%. In some embodiments, Tl is between 0.25% and 0.5%, between 0.25% and 0.45%, between 0.25% and 0.4%, between 0.25% and 0.35%, between 0.25% and 0.3%, between 0.3% and 0.5%, between 0.3% and 0.45%, between 0.3% and 0.4%, between 0.3% and 0.35%, between 0.35% and 0.45%, between 0.35% and 0.4%, between 0.4% and 0.5%, between 0.4% and 0.45%, or between 0.45% and 0.5%.

[0073] In some embodiments, patients are selected for IgG-RT if the frequency of IgG antibody clones having an IGHV3-23 heavy chain V region is greater than a threshold m (Tm), where Tm is at least 6%. In some embodiments, Tm is 6%, 7%, 8%, 9%, or 10%. In some embodiments, Tm is between 6% and 10%, between 6% and 9%, between 6% and 8%, between 6% and 7%, between 7% and 10%, between 7% and 8%, between 8% and 10%, between 8% and 9%, or between 9% and 10%.

[0074] In some embodiments, patients are selected for IgG-RT if the frequency of IgG antibody clones having an IGHV4-34 heavy chain V region is greater than a threshold n (Tn), where Tn is at least 6%. In some embodiments, Tn is 6%, 7%, 8%, 9%, or 10%. In some embodiments, Tn is between 6% and 10%, between 6% and 9%, between 6% and 8%, between 6% and 7%, between 7% and 10%, between 7% and 8%, between 8% and 10%, between 8% and 9%, or between 9% and 10%.

[0075] In some embodiments, patients are selected for IgG-RT if the frequency of IgM antibody clones having an IGHV4-31 heavy chain V region is below a threshold o(T), where T is at most 0.2%. In some embodiments, T is 0.2%, 0.15%, or 0.1%. In some embodiments, T is between 0.1% and 0.2%, between 0.1% and 0.15%, or between 0.15% and 0.2%.

[0076] In some embodiments, patients are selected for IgG-RT if the average percent germline identity of IgG V genes is greater than a threshold p (Tp), where Tp is at least 98%. In some embodiments, Tp is 98%, 98.5%, or 99%. In some embodiments, Tp is between 98% and 99%, between 98% and 98.5%, or between 98.5% and 99%.

[0077] In some embodiments, patients are selected for IgG-RT if the average percent germline identity of IgG J genes is greater than a threshold q (Tq), where Tq is at least 98%. In some embodiments, Tq is 98%, 98.5%, or 99%. In some embodiments, Tq is between 98% and 99%, between 98% and 98.5%, or between 98.5% and 99%.

[0078] In some embodiments, patients are selected for IgG-RT if the median somatic nucleotide mutation frequency in the FR1, CDR1, FR2, or CDR2 region of IgG is below a threshold value (Tr), where Tr is at most 1. In some embodiments, Tr is 1, 0.8, or 0.6. In some embodiments, Tr is a number between 0.6 and 1, between 0.6 and 0.8, or between 0.8 and 1.

[0079] In some embodiments, patients are selected for IgG-RT if the median somatic nucleotide mutation frequency in the FR3 region of IgG is less than Ts, where Ts is up to 4. In some embodiments, Ts is 4, 3.5, 3, 2.5, or 2. In some embodiments, Ts is a number between 2 and 4, between 2 and 3.5, between 2 and 3, between 2 and 2.5, between 2.5 and 4, between 2.5 and 3.5, between 2.5 and 3, between 3 and 4, between 3 and 3.5, or between 3.5 and 4.

[0080] In some embodiments, patients are selected for IgG-RT when one or more of (g) through (s) are met: (g) the number of IgG clones in the BCR repertoire is greater than Tg, wherein Tg is at least 600; (h) the diversity index value of the IgM antibody clones is greater than Th, wherein Th is at least 250; (i) the total frequency of the 10 to 30 most frequent IgG antibody clones in the BCR repertoire is less than Ti, where Ti is at most 30%; (j) the total frequency of the 10 to 30 most frequent IgM antibody clones in the BCR repertoire is less than Tj, wherein Tj is at most 7%; (k) the frequency of IgG antibody clones having the IGHV4-30-2 heavy chain V region is less than Tk, wherein Tk is at most 0.3%; (l) the frequency of IgG antibody clones having the IGHV4-30-4 heavy chain V region is less than Tl, wherein Tl is at most 0.5%; (m) the frequency of IgG antibody clones having the IGHV3-23 heavy chain V region is higher than Tm, wherein Tm is at least 6%; (n) the frequency of IgG antibody clones having the IGHV4-34 heavy chain V region is higher than Tn, wherein Tn is at least 6%; (o) the frequency of IgM antibody clones having the IGHV4-31 heavy chain V region is less than To, wherein To is at most 0.2%; (p) the average percent germline identity of the IgG V genes is greater than Tp, where Tp is at least 98%; (q) the average percent germline identity of the IgG J genes is greater than Tq, where Tq is at least 98%; (r) the median somatic nucleotide mutation frequency in the FR1, CDR1, FR2, or CDR2 region of IgG is less than Tr, where Tr is at most 1; and (s) The median somatic nucleotide mutation frequency in the FR3 region of IgG is less than Ts, where Ts is at most 4.

[0081] In some embodiments, patients are selected for IgG-RT when one criterion selected from (g)-(s) is met. In some embodiments, patients are selected for IgG-RT when two criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when three criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when four criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when five criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when six criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when seven criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when eight criteria selected from (g)-(s) are met. In some embodiments, patients are selected for IgG-RT when nine criteria selected from (g)-(s) are met. In some embodiments, a patient is selected for IgG-RT when 10 criteria selected from (g)-(s) are met. In some embodiments, a patient is selected for IgG-RT when 11 criteria selected from (g)-(s) are met. In some embodiments, a patient is selected for IgG-RT when 12 criteria selected from (g)-(s) are met. In some embodiments, a patient is selected for IgG-RT when 13 criteria selected from (g)-(s) are met.

[0082] In some embodiments, patients are selected for IgG-RT if at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, or at least twelve criteria selected from (g) through (s) are met.

[0083] In some embodiments, the methods use a patient selected for having less than 5 g / L of serum IgG and more than 40 / μL of peripheral B cells. In some embodiments, the patient is selected for having less than 4.5 g / L of serum IgG and more than 35 / μL of peripheral B cells. In some embodiments, the patient is selected for having less than 4 g / L of serum IgG and more than 30 / μL of peripheral B cells.

[0084] 6.2 Providing information on whether a patient requires IgG-RT The methods disclosed herein include providing information regarding whether a patient requires IgG-RT. In some embodiments, the information is provided to the patient or the patient's guardian. In some embodiments, the information is provided to a medical professional. In some embodiments, the information is provided as a report. In some embodiments, the information is provided online, for example, on a website or by email. In some embodiments, the information is provided on a screen.

[0085] In some embodiments, the recipient of the information decides whether to treat the patient with IgG-RT or whether to receive IgG-RT. Thus, in some embodiments, the method further comprises treating the patient with IgG-RT. In some embodiments, the method comprises treating the patient with a therapy other than IgG-RT.

[0086] 6.2 Treatment method Another aspect of the present disclosure relates to treating patients with hypogammaglobulinemia. In some embodiments, the treatment method includes determining whether to treat the patient with IgG-RT. In some embodiments, the patient has not been treated with IgG-RT.

[0087] In some embodiments, the method includes analyzing the patient's B cell repertoire as described herein. In some embodiments, the method includes treatment with IgG-RT only if the patient has been selected for IgG-RT using the methods described herein. In some embodiments, the method includes an additional treatment known to be effective for treating hypogammaglobulinemia. In some embodiments, the method includes a treatment other than IgG-RT when the patient has not been selected for IgG-RT.

[0088] 6.2 Diagnostic Products and Methods In yet another aspect, the present disclosure provides a diagnostic product for selecting hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT). In some embodiments, the diagnostic product is designed to use the methods for selecting hypogammaglobulinemic patients for treatment with IgG-RT described herein. In some embodiments, the diagnostic product is stored on a non-transitory computer-readable medium. In some embodiments, the diagnostic product is a set of trained parameters for a machine learning (ML) or artificial intelligence (AI) model.

[0089] In some embodiments, the diagnostic product is manufactured by a process comprising: (1) obtaining sequence information of a plurality of patients with hypogammaglobulinemia, wherein the sequence information of each patient comprises the sequences of at least 10,000 transcripts from a sample of said patient comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) characterizing the B cell receptor (BCR) repertoire of each of said patients by identifying antibody clones based on said sequence information; (3) obtaining a training dataset comprising a plurality of training examples, each training example corresponding to the BCR repertoire of an individual patient; and (a) one or more characteristics associated with the BCR repertoire of the individual patient; and (b) diagnosing said hypogammaglobulinemic patient as to whether said individual patient requires IgG-RT, optionally said diagnosis being based on information related to serum IgG levels or susceptibility to infection; and (4) numerically coding the training examples in the training dataset, including numerically coding the one or more characteristics associated with the BCR repertoire of the individual patient and numerically coding the diagnosis of the hypogammaglobulinemic patient as to whether the individual patient requires IgG-RT; (5) For a diagnostic model comprising a neural network having a plurality of layers, each layer having a plurality of parameters, the layers comprising an input layer for receiving the one or more numerically encoded features of the BCR repertoire, and an output layer indicative of the likelihood of need for IgG-RT, during one or more iterations of a training process: (a) applying the parameters of the neural network to a set of training examples for a current iteration to generate an estimated likelihood for the set of training examples; (b) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (c) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (d) stopping the backpropagation after the loss function meets a criterion; and (6) storing the updated set of parameters for the layer of the diagnostic model in the computer-readable storage medium.

[0090] In some embodiments, in step (3)(a), the one or more characteristics associated with the BCR repertoire of the individual patient are selected from 1) to 7), 1) the number or abundance of individual IgG clones in the BCR repertoire; 2) an IgM antibody clonal diversity index value calculated by measuring the number and abundance of individual IgM antibody clones within the BCR repertoire; 3) the total frequency of the top 10, top 11, top 12, top 13, top 14, top 15, top 16, top 17, top 18, top 19, top 20, top 21, top 22, top 23, top 24, top 25, top 26, top 27, top 28, top 29, or top 30 IgG antibody clones that are most frequently found in the BCR repertoire; 4) the total frequency of the top 10, top 11, top 12, top 13, top 14, top 15, top 16, top 17, top 18, top 19, top 20, top 21, top 22, top 23, top 24, top 25, top 26, top 27, top 28, top 29, or top 30 IgM antibody clones that are most frequently found in the BCR repertoire; 5) variable region (V) gene usage in the antibody clone, optionally wherein the V gene is selected from the group consisting of an IGHV4-30-2 heavy chain V gene, an IGHV4-30-4 heavy chain V gene, an IGHV3-23 heavy chain V gene, an IGHV4-34 heavy chain V gene, and an IGHV4-31 heavy chain V gene; 6) the average percent germline identity measured by comparing the variable (V) or joining (J) regions in the BCR repertoire to the corresponding germline sequences; and 7) the median somatic nucleotide mutation frequency in the V region in the BCR repertoire, optionally the V region comprising FR1, CDR1, FR2, CDR2 and / or FR3 regions; is selected from.

[0091] In some embodiments, each of the one or more features associated with an individual patient's BCR repertoire selected from 1)-7) is numerically coded as a scalar, vector, or tensor. In some embodiments, one or any combination of the one or more features associated with an individual patient's BCR repertoire is input to the neural network as one or more scalars concatenated into a vector or tensor.

[0092] In some embodiments, the hypogammaglobulinemic patient diagnosis of whether or not the individual patient required IgG-RT is numerically coded as a categorical variable. In some embodiments, the categorical variable is a binary variable that is coded as a non-zero value (e.g., a value of "1") if IgG-RT was required for the individual patient and as a value of 0 if IgG-RT was not required for the individual patient.

[0093] In some embodiments, the hypogammaglobulinemic patient diagnosis of whether an individual patient required IgG-RT is numerically coded as a continuous variable indicating the degree to which IgG-RT was required for the individual patient.

[0094] In some embodiments, the loss function for the current iteration of the training process is one or more of the L norm, L norm, L infinity norm, cross entropy loss. In some embodiments, the cross entropy loss is calculated as follows:

number

number

number

number

[0095] In some embodiments, backpropagation of the current iteration is performed by computing the gradient of the iteration's computed loss with respect to the parameter space of the neural network, and updating the previous set of parameters by the coefficients of the computed gradient.

[0096] In some embodiments, the architecture of the neural network is configured as an artificial neural network (ANN), a feedforward neural network, a deep neural network (DNN), a recurrent neural network (RNN), a transformer neural network with one or more attention layers, etc. In some embodiments, the number of parameters of the neural network is greater than 1,000 parameters, 10,000 parameters, 100,000 parameters, 1 million parameters, 1 billion parameters, 10 billion parameters, 100 billion parameters, 1 trillion parameters.

[0097] In another aspect, the present disclosure provides a diagnostic product for selecting hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT), wherein said diagnostic product is stored on a non-transitory computer readable medium and comprises: (1) obtaining sequence information of a plurality of patients with hypogammaglobulinemia, wherein the sequence information of each patient comprises the sequences of at least 10,000 transcripts from a sample of said patient comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) obtaining a training dataset comprising a plurality of training examples, wherein each training example comprises sequence information of said individual patient and a diagnosis of said individual patient as to whether said individual patient requires IgG-RT, optionally wherein said diagnosis is based on information related to serum IgG levels or susceptibility to infection; (3) numerically encoding the training examples in the training dataset, including numerically encoding the sequence information of the individual patient and numerically encoding the individual patient's diagnosis of whether the individual patient requires IgG-RT; (4) For a diagnostic model including a neural network having a plurality of layers, each layer having a plurality of parameters, the layer including an input layer for receiving the numerically coded sequence information and an output layer indicative of the likelihood of need for IgG-RT, during one or more iterations of a training process: (a) applying the parameters of the neural network to a set of training examples for a current iteration to generate an estimated likelihood for the set of training examples; (b) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (c) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (d) stopping the backpropagation after the loss function meets a criterion; and (5) storing an updated set of parameters for the layer of the diagnostic model in the computer-readable storage medium; The present invention provides a diagnostic product manufactured by a process comprising:

[0098] In some embodiments, the sequence information for an individual patient is numerically coded as a scalar, vector, or tensor.

[0099] In some embodiments, the hypogammaglobulinemic patient diagnosis of whether or not the individual patient required IgG-RT is numerically coded as a categorical variable. In some embodiments, the categorical variable is a binary variable that is coded as a non-zero value (e.g., a value of "1") if IgG-RT was required for the individual patient and as a value of 0 if IgG-RT was not required for the individual patient.

[0100] In some embodiments, the hypogammaglobulinemic patient diagnosis of whether an individual patient required IgG-RT is numerically coded as a continuous variable indicating the degree to which IgG-RT was required for the individual patient.

[0101] In some embodiments, the loss function for the current iteration of the training process is one or more of the L norm, L norm, L infinity norm, cross entropy loss. In some embodiments, the cross entropy loss is calculated as follows:

number

number

number

number

[0102] In some embodiments, backpropagation of the current iteration is performed by computing the gradient of the iteration's computed loss with respect to the parameter space of the neural network, and updating the previous set of parameters by the coefficients of the computed gradient.

[0103] In some embodiments, the architecture of the neural network is configured as an artificial neural network (ANN), a feedforward neural network, a deep neural network (DNN), a recurrent neural network (RNN), a transformer neural network with one or more attention layers, etc. In some embodiments, the number of parameters of the neural network is greater than 1,000 parameters, 10,000 parameters, 100,000 parameters, 1 million parameters, 1 billion parameters, 10 billion parameters, 100 billion parameters, 1 trillion parameters.

[0104] In one aspect, the present disclosure provides methods of using the diagnostic products to select hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT).

[0105] In some embodiments, the method comprises: (1) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, wherein each of the transcripts encodes an IgG or IgM heavy chain or a portion thereof; (2) providing the sequence information or information related to the patient's B-cell receptor (BCR) repertoire obtained by processing the sequence information to a diagnostic product disclosed herein and operating the diagnostic product; (3) obtaining from the diagnostic product information regarding whether the patient requires IgG-RT; Includes:

[0106] In some embodiments, trained parameters of a neural network stored in the diagnostic product are applied to sequence information or information related to the B cell receptor (BCR) repertoire to generate a likelihood that a subject will or will not require IgG-RT.

[0107] In some embodiments, the step (2) of characterizing the B cell receptor (BCR) repertoire comprises (a) to (f): (a) determining the number or abundance of individual antibody clones in the BCR repertoire; (b) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (c) selecting at least 10 but not more than 30 antibody clones that are most frequent in the BCR repertoire and calculating the total frequency of the 10 to 30 frequent antibody clones; (d) determining variable region gene (V region) usage in said antibody clones, optionally wherein said V is selected from the group consisting of an IGHV4-30-2 heavy chain V region gene, an IGHV4-30-4 heavy chain V region gene, an IGHV3-23 heavy chain V region gene, an IGHV4-34 heavy chain V region gene, and an IGHV4-31 heavy chain V region gene; (e) determining percent germline identity by comparing the V or J regions in the BCR repertoire with corresponding germline sequences; and (f) determining somatic mutation frequencies at different regions along the V regions; The method comprises analyzing the BCR repertoire by one or more steps selected from:

[0108] In some embodiments, the method further comprises treating the patient with IgG-RT if the patient is selected for IgG-RT.

[0109] In yet another aspect, the present disclosure provides methods of diagnosing a hypogammaglobulinemic patient for treatment with immune globulin replacement therapy (IgG-RT). In some embodiments, the method comprises: (1) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, each transcript encoding an IgG or IgM heavy chain or a portion thereof; (2) receiving the patient's sequence information or the patient's BCR repertoire and obtaining parameters for a trained neural network configured to generate a likelihood that the patient will or will not require IgG-RT; (3) providing the sequence information or information related to the patient's B-cell receptor (BCR) repertoire obtained by processing the sequence information to a trained neural network as disclosed herein; (4) obtaining information from the trained neural network regarding whether the patient requires IgG-RT.

[0110] In some embodiments, the neural network is trained by a method comprising: (1) obtaining sequence information of a plurality of patients with hypogammaglobulinemia, wherein the sequence information of each patient comprises the sequences of at least 10,000 transcripts from a sample of said patient comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) characterizing the B cell receptor (BCR) repertoire of each of said patients by identifying antibody clones based on said sequence information; (3) obtaining a training dataset comprising a plurality of training examples, each training example corresponding to the BCR repertoire of an individual patient; and (a) one or more characteristics associated with the BCR repertoire of the individual patient; and (b) diagnosing said hypogammaglobulinemic patient as to whether said individual patient requires IgG-RT, optionally said diagnosis being based on information related to serum IgG levels or susceptibility to infection; and (4) numerically coding the training examples in the training dataset, including numerically coding the one or more characteristics associated with the BCR repertoire of the individual patient and numerically coding the diagnosis of the hypogammaglobulinemic patient as to whether the individual patient requires IgG-RT; (5) For a diagnostic model comprising a neural network having a plurality of layers, each layer having a plurality of parameters, the layer comprising an input layer for receiving the one or more numerically encoded features of the BCR repertoire, and an output layer indicative of the likelihood of need for IgG-RT, during one or more iterations of a training process: (a) applying the parameters of the neural network to a set of training examples for a current iteration to generate an estimated likelihood for the set of training examples; (b) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (c) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (d) stopping the backpropagation after the loss function satisfies a criterion.

[0111] In some embodiments, the neural network is trained by a process that includes: (1) obtaining sequence information of a plurality of patients with hypogammaglobulinemia, wherein the sequence information of each patient comprises the sequences of at least 10,000 transcripts from a sample of said patient comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) obtaining a training dataset comprising a plurality of training examples, wherein each training example comprises sequence information of said individual patient and a diagnosis of said individual patient as to whether said individual patient requires IgG-RT, optionally wherein said diagnosis is based on information related to serum IgG levels or susceptibility to infection; (3) numerically encoding the training examples in the training dataset, including numerically encoding the sequence information of the individual patient and numerically encoding the individual patient's diagnosis of whether the individual patient requires IgG-RT; (4) For a diagnostic model including a neural network having a plurality of layers, each layer having a plurality of parameters, the layer including an input layer for receiving the numerically coded sequence information and an output layer indicative of the likelihood of need for IgG-RT, during one or more iterations of a training process: (a) applying the parameters of the neural network to a set of training examples for a current iteration to generate an estimated likelihood for the set of training examples; (b) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (c) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (d) stopping the backpropagation after the loss function satisfies a criterion.

[0112] By training and deploying a neural network and a diagnostic product that stores its parameters, it is possible to learn patterns and characteristics of the BCR repertoire and / or sequence information that indicates the maturity or immaturity of a patient's immune cells and whether the patient requires IgG-RT to treat hypogammaglobulinemia. [Example]

[0113] 7. Working Example 6.2 Experimental methods to identify differences in hypogammaglobulinemic patients who did and did not require immunoglobulin replacement therapy Sample collection Patients were identified from the Adult Outpatient Immunodeficiency Clinic at the University of Freiburg as having reduced serum IgG levels (<4 g / L) and residual peripheral B cells >40 / μL. For patients requiring IgG-RT (those with recurrent respiratory tract infections, n = 15), hypogammaglobulinemia was assessed using retrospective data from the time of diagnosis (before starting regular IgG-RT). Hypogammaglobulinemic patients (n = 10) who did not have recurrent respiratory tract infections were not prescribed IgG-RT. Patients' infectious history, other noninfectious diagnoses, and ability to respond to vaccines are shown in Table 1.

[0114] Participants provided blood samples after signing written informed consent. PBMCs from the provided blood samples were isolated using Ficoll / Pancoll density gradient centrifugation under sterile conditions according to standard protocols. Recovered PBMCs (9–17 × 10) were frozen in freezing medium (heat-inactivated 90% fetal bovine serum (FBS) + 10% dimethyl sulfoxide (DMSO)). 6 The cultured cells (cells / ml) were stored in liquid nitrogen until further processing.

[0115] Flow cytometry Red blood cells from 500 μl of whole blood were lysed with ammonium chloride for 10 min at 4°C, washed twice with phosphate-buffered saline (PBS) + 2% FBS, and stained with anti-CD19 (APC-Cy7, HIB19, Biolegend), anti-CD27 (BV421, M-T271, Biolegend), anti-IgD (PE, IA6-2, Biolegend), anti-IgA (FITC, goat IgG, Southern Biotech), and anti-IgG (AF700, G18-145, BD Biosciences) for 20 min at room temperature. Cells were then fixed (Optilyse B, Beckman Coulter) for 20 min at room temperature, followed by another wash step with PBS + 2% FBS. Stained cells were measured using a Navios Flow Cytometer (Beckman-Coulter) and analyzed using Kaluza Analysis Software (Beckman-Coulter).

[0116] Antibody Repertoire Sequencing Harvested PBMCs were thawed in medium (RPMI + 10% FBS) and counted using a Cellometer K2 (Nexcelom). Cells were pelleted by centrifugation, and RNA was extracted using the NucleoSpin RNA Plus kit (Macherey-Nagel) according to the manufacturer's instructions. To amplify heavy chain variable regions for deep sequencing, tailed-end RT-PCR was performed on the extracted RNA. A pool of variable region primers with Illumina adapters was used at the 5' end, and constant region primers (for IgG or IgM) with sample-specific index sequences and Illumina adapters were used at the 3' end (Table 2). IgG and IgM sequences were amplified in separate reactions. PCR products were run on an agarose gel, extracted, purified, and quantified using the KAPA quantitative PCR Illumina Library Quantification Kit (1069, Roche). Libraries were sequenced on a MiSeq (Illumina) at a library concentration of 9 pM using 255 cycles of forward reads and 255 cycles of reverse reads as previously described (see Table 2 for sequencing primers). Sequencing data are available in the Short Read Archive under project identifier PRJNA876301.

[0117] Antibody sequence analysis Antibody repertoire libraries were sequenced to an average of 28,901 reads (range: 13,064-45,080 reads). Sequence analysis was performed. Briefly, the expected error count (E) of a read was calculated from its Phred score and reads with E > 2 were discarded. After error filtering, up to 15,000 reads were randomly sampled from each sample for further analysis. Applicants confirmed that our findings were consistent across multiple rounds of random read sampling (data not shown). IMGT immunoglobulin sequences were processed to generate position-specific sequence matrices (PSSMs) for each framework / CDR junction. These PSSMs were used to identify framework / CDR junctions for each nucleotide sequence. Sequences were then translated using a Python script. Reads were required to have valid predicted CDR3 sequences. Applicants then conservatively defined antibody "clones" and combined unique sequences if they had a single amino acid difference for a CDR3H that was 5-6 amino acids long, or 1-2 amino acid differences for a CDR3H that was longer than 6 amino acids. Only clones with at least two sequencing reads were included in the analysis.

[0118] UBLAST was performed using the nucleotide sequence as a query and V and J gene sequences from the IMGT database as a reference sequence. The UBLAST alignment with the lowest E-value was used to assign V and J gene families and calculate percent identity to the germline. The IgG sample from patient CVID-1712-01 had poor sequence quality and was excluded from the analysis.

[0119] Antibody diversity index The antibody diversity index was calculated using the diversity function in the tcR package (version 2.3.2) in R version 4.1.2. The true diversity of an antibody repertoire X refers to the effective richness of that population, i.e., the number of equally common antibody clones that would be required to produce a repertoire with the same overall diversity as X. This value increases with the number of antibody clones in the repertoire, as well as with the evenness with which these clones are distributed.

[0120] Correlation analysis The data used for the correlation analysis are shown in Table 1. Pearson correlation analysis was performed using the cor function in the corrplot package (version 0.92) using the "pairwise.complete.obs" option in R version 4.1.2. Correlations with p ≤ 0.05 were considered significant.

[0121] Variable (V) gene usage and mutation frequency To identify antibody V gene identities, sequencing fasta files were mapped to the IMGT human V gene reference sequence (release 202243-1, October 24, 2022) using USEARCH version v8.1.1916 M_i86linux®64 (options: -usearch_local-mismatch-1-id 0.5-evalue 1e-3). The IMGT antibody numbering system was used to identify CDR and framework regions along the V genes (also used to determine CDR3H length). Principal component analysis (PCA) was performed using log2-transformed V gene usage. The Wilcoxon rank-sum test was used to compare V gene usage between donors who required IgG-RT and those who did not. The Benjamini-Hochberg method was used to adjust P values ​​for the number of V genes tested. The number of mismatches along the V genes was tabulated using a custom Perl script and visualized using ggplot2 in R. The first 21 nucleotides (7 amino acids) of the V gene were PCR primer binding sites for preparing antibody sequencing libraries, and mutations in this region could not be accurately measured and were therefore excluded from the V gene mutation frequency analysis.

[0122] 6.2 Differences between patients with hypogammaglobulinemia who required immunoglobulin replacement therapy and those who did not Patient cohort Based on susceptibility to infection, 25 patients with low IgG serum concentrations were recruited, of whom 15 required IgG-RT (3 men; 12 women), and 10 did not (8 men; 2 women) (Table 1). On average, patients who required IgG-RT had serum IgG levels of 1.86 g / L (standard deviation, SD = 1.31), IgM levels of 0.24 g / L (SD = 0.15), and IgA levels of 0.10 g / L (SD = 0.072) before IgG-RT, whereas patients who did not require IgG-RT had serum levels of 2.69 g / L (SD = 1.11), IgM levels of 0.39 g / L (SD = 0.31), and IgA levels of 0.71 g / L (SD = 0.64) (Figure 1A). Serum IgA titers were significantly different between the two groups (p = 0.0019). Patients who required IgG-RT had similar amounts of CD19 compared with patients who did not require IgG-RT (208.9 cells / μl, SD=279.5; p=0.24). + The patients had a high B cell count (mean = 221.3 cells / µL, SD = 146.1). Vaccine responses to various pathogens (e.g., tetanus, diphtheria, and pneumococcal polysaccharide) were observed in most patients who did not require IgG-RT compared with patients who did require IgG-RT. Furthermore, autoimmune manifestations, chronic infections, and other complications were more common in patients who required IgG-RT (Table 1).

[0123] Antibody Repertoire Sequencing IgG and IgM antibody repertoire sequencing of heavy chain immunoglobulins for both patient cohorts from isolated PBMCs was performed (the IgA repertoire could not be investigated in this study due to low or absent IgA memory B cell numbers in most patients requiring IgG-RT; Figure 1A, Table 1). Antibody "clones" were conservatively defined as unique sequences with one amino acid difference within the CDR3H (complementarity-determining region 3H) of 5–6 amino acids in length, or one to two amino acid differences relative to the CDR3H of longer than 6 amino acids in length. Only clones with at least two sequencing reads were included in the analysis. Interestingly, patients who required IgG-RT had a significantly higher number of IgG clones (mean = 1,310 clones) than patients who did not require IgG-RT (mean = 483 clones; p = 0.0015) (Figure 1B). On the other hand, there was no significant difference in the number of IgM clones between those who required IgG-RT (mean = 2,760 clones) and those who did not (mean = 2,733 clones; p = 0.96).

[0124] To further examine IgG and IgM antibody repertoires, the true diversity index was measured, which takes into account the abundance of individual antibody clones in addition to the number of clones. The true diversity of an antibody repertoire X refers to the effective richness of that population, i.e., the number of equally common antibody clones that would be required to produce a repertoire with the same overall diversity as X. This value increases with the number of antibody clones in the repertoire as well as with the evenness with which these clones are distributed. Compared to patients who did not require IgG-RT, patients who required IgG-RT had a significantly higher IgM diversity index (p=8.5×10 -5 ) (Figure 1C).

[0125] Of the donors who did not require IgG-RT, one donor had an IgG titer of 4.18 g / L and an additional donor had an IgM titer of 1.2 g / L (Figure 1A, Table 1). To ensure that these donors with higher antibody titers were not driving differences in antibody clone number and diversity, we removed these donors from the data set (Figure 4A) and repeated the above analysis. We observed the same differences, except now donors who required IgG-RT had significantly higher IgG clone numbers (p=0.0016; Figure 4B) and a higher IgM diversity index (p=4.1×10 -6 ; Figure 4C).

[0126] Visualization of the frequencies of the top 20 antibody clones showed that patients requiring IgG-RT tended to have smaller oligoclonal IgG and IgM repertoires (Figure 1D, Figure 1E). On average, the top 20 IgG clones comprised 19.5% and 42.1% of the total repertoire for patients who required IgG-RT and patients who did not, respectively, indicating lower IgG oligoclonality for the former cohort (p = 0.0015). Similarly, the top 20 IgM clones accounted for an average of 2.65% and 7.06% of the repertoire for patients who required IgG-RT and patients who did not, respectively, indicating lower IgM oligoclonality for the former cohort (p = 0.0014).

[0127] Applicants further investigated the distribution of CDR3H amino acid sequence lengths, another feature that may provide insight into the composition of the antibody repertoire. However, both patient cohorts had a normal distribution of heavy chain CDR3 lengths with a median of 15 amino acids for both IgG and IgM (Figure 1F).

[0128] Together, these data indicate that patients who required IgG-RT had more IgG clones, a higher IgM diversity index, and lower IgG and IgM oligoclonality, consistent with a more diverse antibody repertoire.

[0129] Correlation between antibody repertoire and immune signature Next, we investigated the interactions between different features of the antibody repertoire and various immune parameters. The frequencies of different B cell subtypes were measured by flow cytometry for both patient cohorts. We then performed a full-component correlation analysis of antibody titers, clonal numbers, diversity, and abundance of different B cell subtypes for patients who required IgG-RT and those who did not (Figures 2A-J, Figure 5; Table 1).

[0130] For patients who did not require IgG-RT, IgG diversity was significantly higher in CD19+ B cells (Pearson correlation coefficient, r = 0.91, p = 0.00028) and IgD+CD27+ B cells (r = 0.98, p = 1.34 × 10 -6 In the same patient cohort, IgM diversity was also positively correlated with the frequency of CD19+ B cells (r = 0.76, p = 0.011) and IgD+CD27+ B cells (r = 0.8, p = 0.0051) (Figures 2A, 2E, 2F). IgG diversity was negatively correlated with the frequency of IgD+CD27- naive B cells (r = -0.82, p = 0.0039) (Figures 2A, 2G). These correlations between IgG and IgM diversity and B cell frequency were not observed in patients who required IgG-RT (Figure 2B).

[0131] For patients who required IgG-RT, IgG titers correlated with the number of IgM clones (r = 0.64, p = 0.011) (Figure 2B, Figure 2H). IgG titers also correlated with the frequencies of IgA+CD27+ B cells (r = 0.75, p = 0.013) and IgD-CD27+ memory B cells (r = 0.69, p = 0.0048) (Figure 2b, Figure 2i, Figure 2j). These correlations were not observed in patients who did not require IgG-RT (Figure 2A).

[0132] V and J gene diversity V(D)J (variable, diversity, joining) recombination, which constructs antibody gene segments during B cell development, contributes to the vast combinatorial diversity of antibodies. We evaluated whether V(D)J diversity differed between patients who required IgG-RT and those who did not. For IgG and IgM, both patient cohorts showed diverse V and J gene usage (Figures 3A, 3B, 6A, and 6B). Interestingly, principal component analysis (PCA) of IgG V gene frequencies revealed that patients clustered based on their need for IgG-RT. Principal component 1 (PC1) explained 15.55% of the variance in V gene usage and separated patients who required IgG-RT from those who did not (Figure 6C). PCA of IgM V gene usage showed less clustering of the patient cohorts (Figure 6D). Next, we compared V gene frequencies between patients who required IgG-RT and those who did not. Compared to patients who did not require IgG-RT, patients who required IgG-RT had fewer IgG antibody clones with IGHV4-30-2 and IGHV4-30-4 heavy chain V genes (Benjamini-Hochberg adjusted p-value = 0.04) (Figure 3C). Patients who required IgG-RT also had an increased number of antibody clones with IGHV3-23 and IGHV4-34 V genes (adjusted p = 0.04) (Figure 3C). Patients who did not require IgG-RT had, on average, 4.53% IGHV4-34 clones, consistent with the 3-9% prevalence of the gene in adult B lymphocytes, whereas patients who required IgG-RT had, on average, 11.27% IGHV4-34 antibody clones (Figure 3C). Notably, antibodies with the IGHV4-34 V gene have been shown to be autoreactive and are more common in the naive B cell repertoire than in memory B cells. We also examined differences in IgM V gene usage. Patients who required IgG-RT had fewer IgM clones with IGHV4-31 V genes (adjusted p = 0.04) (Figure 3D). Finally, we examined J gene usage and did not observe any significant differences between patient cohorts for either IgG or IgM.

[0133] Somatic hypermutation, the process by which point mutations accumulate across the antibody V(D)J region, further contributes to antibody diversity. Somatic hypermutation is also an important tool for generating high-affinity antibodies. We measured the percent nucleotide identity of antibody heavy chain V and J genes to their respective germline sequences. Specifically, to identify the identity of the germline V and J genes, we used UBLAST to map the antibody nucleotide sequences to the IMGT human V and J gene reference sequence (release 202243-1, October 24, 2022). The UBLAST alignment with the lowest E-value was used to assign the V and J gene family and calculate the percent identity to the germline sequence. Interestingly, compared with patients who did not require IgG-RT, patients who required IgG-RT had significantly higher percent IgG V and J gene germline identity (p ≤ 0.0001; Figure 3E). For IgM, the percent germline identity of V genes was significantly lower in those requiring IgG-RT (p ≤ 0.0001; Figure 3F), but the mean difference was small (98.15% in those not requiring IgG-RT vs. 98.39% in those requiring IgG-RT). When donors with higher IgG / IgM titers were removed from the dataset, the difference in percent germline identity of IgG V and J genes remained significant, suggesting that the observations were not driven by the highest titer donors (Figures 7A and 7B). To further investigate the differences in V gene mutations between the two patient cohorts, we measured mutation frequencies in different regions along the V genes, including the framework regions (FR1, FR2, FR3) and complementarity-determining regions (CDR1, CDR2). The patient cohort who required IgG-RT had significantly (p≦0.05) lower mutation frequencies across all V gene regions at both the nucleotide (Figure 3G, Figure 7C) and estimated protein (Figure 7D, Figure 7E) levels for IgG, but not for IgM. Visualization of mutation frequencies along the most common V genes further demonstrated a lower IgG V gene mutation rate in patients who required IgG-RT (Figures 8, 9).

[0134] Finally, we measured the frequency of somatic hypermutations along the V gene IGHV4-34, which increased with IgG use in donors who required IgG-RT. Compared with patients who did not require IgG-RT, patients who required IgG-RT had lower somatic hypermutations along IGHV4-34 (Figure 10A). Previous studies have shown that the autoreactivity of IGHV4-34 antibodies is mediated by a hydrophobic patch within the framework 1 region and that somatic hypermutations within this region can eliminate autoreactivity. However, there was no significant difference in mutation frequency in the hydrophobic patch (AVY residues) when comparing the two cohorts (Figure 10B).

[0135] Overall, these data indicate that IgG hypogammaglobulinemia patients who required IgG-RT and those who did not required IgG-RT had antibody repertoires with different V gene diversity. Patients who required IgG-RT showed a higher use of V genes related to the naive antibody repertoire, suggesting less somatic hypermutation in their IgG clones and a smaller mature antibody repertoire, possibly making these patients more susceptible to infection.

[0136] conclusion The decision to treat patients with hypogammaglobulinemia with IgG-RT can be difficult because both IgG levels and susceptibility to infection vary among patients. IgG levels do not necessarily predict a patient's susceptibility to infection, and in some cases, IgG-RT is recommended for patients with asymptomatic hypogammaglobulinemia due to the potential risk of severe infection. Furthermore, patients with both symptomatic and asymptomatic hypogammaglobulinemia can respond well to tetanus vaccination, but their diphtheria response is often impaired. In fact, most patients in this study had a positive response to tetanus vaccination before IgG-RT was initiated for those who required it, but many did not respond to diphtheria (Table 1). Furthermore, eight of nine patients who were vaccinated with pneumococcal polysaccharide and did not require IgG-RT had a positive response, whereas only one of four patients who required IgG-RT responded.

[0137] Hypogammaglobulinemic patients who required IgG-RT and those who did not had multiple differences in their peripheral B cell receptor repertoires. Patients who required IgG-RT had more IgG antibody clones, a higher IgM diversity index, and fewer oligoclonal IgG and IgM repertoires. Their IgG clones showed distinct heavy chain V gene usage, with a higher frequency of sequences with V genes associated with the naive B cell repertoire. Their IgG clones were less somatically hypermutated and appeared more similar to germline sequences. Lower levels of clonal antibody proliferation and somatic hypermutation suggest that these infection-susceptible patients have relatively immature B cell receptor repertoires that may be less effective against pathogens. A reduced frequency of somatic hypermutation was similarly found in the B cell receptor repertoires of patients with common variable immunodeficiency (CVID), further suggesting impaired repertoire specification in germinal centers. Interestingly, patients requiring IgG-RT showed increased IGHV4-34 and IGHV3-23 gene usage compared with patients not requiring IgG-RT. Increased IGHV4-34 was observed in CD19-deficient patients, patients with Wiskott-Aldrich syndrome (WAS), and patients with RAG deficiency, indicating its role in autoreactive autoantibodies. Tipton et al. summarized reports of increased IGHV4-34 gene usage in patients with systemic lupus erythematosus and concluded that incomplete tolerance and 9G4 idiotypic autoantibodies are another prominent feature in the repertoire of this disease. The IGHV3-23 gene has been shown to be associated with exposure to self and / or environmental antigens and is relatively abundant in humans. IGHV3-23 gene usage has also been reported in hairy cell leukemia, diffuse large B-cell lymphoma, malaria-naive individuals after immunization with PfSPZ-CVac, HIV patients, and CD21(low) B cells from WAS patients.

[0138] Conversely, hypogammaglobulinemic patients who did not require IgG-RT possessed a relatively expanded antigen-experienced B cell repertoire that may be better adapted to overcome susceptibility to infection. These patients demonstrated higher gene usage of IGHV4-30-2, IGHV4-30-4, and IGHV4-31 compared with patients requiring IgG-RT. Increased IGHV4-30-2 and -4 have also been reported in WAS patients, demonstrating abnormalities in the immune repertoire in both cohorts.

[0139] This study shows that peripheral B cell receptor sequencing can be utilized in the decision-making process for or against the use of IgG-RT in the setting of hypogammaglobulinemia.

[0140] 8. Equivalents and Incorporation by Reference While the present invention has been particularly shown and described with reference to preferred and various alternative embodiments, it will be understood by those skilled in the art that various changes in form and detail can be made therein without departing from the spirit and scope of the invention.

[0141] All references, issued patents, and patent applications cited within the body of this specification are hereby incorporated by reference in their entirety for all purposes. [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4] [Table 1-5] [Table 1-6] Table 2-1 Table 2-2 Table 2-3

Claims

1. 1. A method for selecting a hypogammaglobulinemic patient for immune globulin replacement therapy (IgG-RT), comprising: (1) obtaining sequence information for at least 10,000 transcripts from a patient sample containing B cells, wherein each transcript encodes an IgG or IgM heavy chain or a portion thereof; (2) characterizing the patient's B cell receptor (BCR) repertoire by identifying antibody clones using the sequence information; and (3) (a) to (f): (t) determining the number or abundance of individual antibody clones in said BCR repertoire; (u) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (v) selecting at least 10 but not more than 30 most frequent antibody clones in said BCR repertoire and calculating the total frequency of said 10 to 30 most frequent antibody clones; (w) determining variable region gene (V region) usage in said antibody clones, wherein optionally said V region genes are selected from the group consisting of IGHV4-30-2 heavy chain V region genes, IGHV4-30-4 heavy chain V region genes, IGHV3-23 genes, IGHV4-34 heavy chain V region genes, and IGHV4-31 heavy chain V region genes; (x) determining percent germline identity by comparing the V or J regions in said BCR repertoire with corresponding germline sequences; and (y) determining somatic mutation frequencies at different regions along the V region. analyzing the BCR repertoire by one or more selected from: (4) (g) to (s): (z) the number of IgG clones in said BCR repertoire is greater than Tg, wherein Tg is at least 600; (aa) the diversity index value of the IgM antibody clones is greater than Th, wherein Th is at least 250; (bb) the total frequency of the 10 to 30 most frequent IgG antibody clones in the BCR repertoire is less than Ti, where Ti is at most 30%; (cc) the total frequency of the 10 to 30 most frequent IgM antibody clones in said BCR repertoire is less than Tj, wherein Tj is at most 7%; (dd) the frequency of IgG antibody clones having the IGHV4-30-2 heavy chain V region is less than Tk, wherein Tk is at most 0.3%; (ee) the frequency of IgG antibody clones having the IGHV4-30-4 heavy chain V region is less than Tl, wherein Tl is at most 0.5%; (ff) the frequency of IgG antibody clones having the IGHV3-23 heavy chain V region is higher than Tm, wherein Tm is at least 6%; (gg) the frequency of IgG antibody clones having the IGHV4-34 heavy chain V region is higher than Tn, wherein Tn is at least 6%; (hh) the frequency of IgM antibody clones having the IGHV4-31 heavy chain V region is less than To, wherein To is at most 0.2%; (ii) the average percent germline identity of the IgG V genes is greater than Tp, where Tp is at least 98%; (jj) the average percent germline identity of the IgG J genes is greater than Tq, wherein Tq is at least 98%; (kk) the median somatic nucleotide mutation frequency in the FR1, CDR1, FR2 or CDR2 region of the IgG is less than Tr, where Tr is at most 1; and (ll) the median somatic nucleotide mutation frequency in the FR3 region of IgG is less than Ts, where Ts is at most 4; selecting the patient for IgG-RT when one or more of the following conditions are met: (5) providing information regarding whether the patient is in need of IgG-RT; A method comprising:

2. The method according to claim 1, wherein in step (1), sequence information of at least 50,000 transcripts is obtained.

3. The method according to claim 2, wherein in step (1), sequence information for at least 100,000, 500,000, 1,000,000, or 10,000,000 transcripts is obtained.

4. The method of any one of claims 1 to 3, wherein in step (4), the patient is selected for IgG-RT when (g) is satisfied.

5. The method of any one of claims 1 to 4, wherein in (g), the Tg is 600, 650, 700, 750, 800, 850, 900, 950, 1,000, 1,050, or 1,100.

6. 5. The method of claim 1, wherein in (g), the Tg is between 700 and 1,200, between 700 and 1,100, between 700 and 1,000, between 800 and 1,100, between 800 and 1,000, between 900 and 1,100, between 900 and 1,000, between 800 and 900, or between 700 and 800.

7. The method of any one of claims 1 to 6, wherein in step (4), the patient is selected for IgG-RT when (h) is satisfied.

8. The method according to any one of claims 1 to 7, wherein in (h), Th is 250, 300, 350, 400, 450, 500, 550 or 600.

9. 8. The method of claim 1, wherein in (h), Th is between 250 and 650, between 300 and 650, between 350 and 650, between 400 and 650, between 450 and 650, between 250 and 600, between 300 and 600, between 350 and 600, between 400 and 600, between 450 and 600, between 250 and 550, between 300 and 550, between 350 and 550, between 400 and 550, between 450 and 550, between 250 and 500, between 300 and 500, between 350 and 500, between 400 and 500, or between 450 and 500.

10. The method of any one of claims 1 to 9, wherein in step (4), the patient is selected for IgG-RT when (i) is met.

11. The method according to any one of claims 1 to 10, wherein in (i) Ti is 30%, 25% or 20%.

12. 11. The method of any one of claims 1 to 10, wherein in (i) Ti is between 20% and 35%, between 20% and 30%, between 20% and 25%, or between 25% and 35%, or between 25% and 30%.

13. The method of any one of claims 1 to 12, wherein in step (4), the patient is selected for IgG-RT when (j) is met.

14. The method according to any one of claims 1 to 13, wherein in (j), Tj is 7%, 6.5%, 6%, 5.5%, 5%, 4.5% or 4%.

15. 14. The method of any one of claims 1 to 13, wherein in (j), Tj is between 4% and 7%, between 4% and 6%, between 4% and 5%, between 5% and 7%, between 5% and 6%, or between 6% and 7%.

16. The method of any one of claims 1 to 15, wherein in step (4), the patient is selected for IgG-RT when (k) is satisfied.

17. The method according to any one of claims 1 to 16, wherein in (k), Tk is 0.3%, 0.25%, 0.2% or 0.15%.

18. 17. The method of any one of claims 1 to 16, wherein in (k), Tk is between 0.15% and 0.3%, between 0.2% and 0.3%, between 0.25% and 0.3%, between 0.15% and 0.25%, between 0.15% and 0.2%, between 0.2% and 0.3%, or between 0.25% and 0.3%.

19. The method of any one of claims 1 to 18, wherein in step (4), the patient is selected for IgG-RT when (l) is met.

20. The method according to any one of claims 1 to 19, wherein in (l), Tl is 0.5%, 0.45%, 0.4%, 0.35%, 0.3% or 0.25%.

21. 20. The method of any one of claims 1 to 19, wherein in (l), Tl is between 0.25% and 0.5%, between 0.25% and 0.45%, between 0.25% and 0.4%, between 0.25% and 0.35%, between 0.25% and 0.3%, between 0.3% and 0.5%, between 0.3% and 0.45%, between 0.3% and 0.4%, between 0.3% and 0.35%, between 0.35% and 0.45%, between 0.35% and 0.4%, between 0.4% and 0.5%, between 0.4% and 0.45%, or between 0.45% and 0.5%.

22. The method of any one of claims 1 to 21, wherein in step (4), the patient is selected for IgG-RT when (m) is met.

23. The method according to any one of claims 1 to 22, wherein in (m), Tm is 6%, 7%, 8%, 9% or 10%.

24. The method of any one of claims 1 to 22, wherein in (m), Tm is between 6% and 10%, between 6% and 9%, between 6% and 8%, between 6% and 7%, between 7% and 10%, between 7% and 8%, between 8% and 10%, between 8% and 9%, or between 9% and 10%.

25. The method of any one of claims 1 to 24, wherein in step (4), the patient is selected for IgG-RT when (n) is met.

26. The method according to any one of claims 1 to 25, wherein in (n), Tn is 6%, 7%, 8%, 9% or 10%.

27. 26. The method of any one of claims 1 to 25, wherein in (n), Tn is between 6% and 10%, between 6% and 9%, between 6% and 8%, between 6% and 7%, between 7% and 10%, between 7% and 8%, between 8% and 10%, between 8% and 9%, or between 9% and 10%.

28. The method of any one of claims 1 to 27, wherein in step (4), the patient is selected for IgG-RT when (o) is met.

29. The method according to any one of claims 1 to 28, wherein in (o), To is 0.2%, 0.15%, or 0.1%.

30. 29. The method of any one of claims 1 to 28, wherein in (o), To is between 0.1% and 0.2%, between 0.1% and 0.15%, or between 0.15% and 0.2%.

31. The method of any one of claims 1 to 30, wherein in step (4), the patient is selected for IgG-RT when (p) is met.

32. The method according to any one of claims 1 to 31, wherein in (p), Tp is 98%, 98.5%, or 99%.

33. The method of any one of claims 1 to 31, wherein in (p), Tp is between 98% and 99%, between 98% and 98.5%, or between 98.5% and 99%.

34. The method of any one of claims 1 to 33, wherein in step (4), the patient is selected for IgG-RT when (q) is satisfied.

35. The method according to any one of claims 1 to 34, wherein in (q), Tq is 98%, 98.5% or 99%.

36. 35. The method of any one of claims 1 to 34, wherein in (q), Tq is between 98% and 99%, between 98% and 98.5%, or between 98.5% and 99%.

37. The method of any one of claims 1 to 36, wherein in step (4), the patient is selected for IgG-RT when (r) is satisfied.

38. The method according to any one of claims 1 to 37, wherein in (r), Tr is 1, 0.8 or 0.

6.

39. 38. The method of any one of claims 1 to 37, wherein in (r), Tr is between 0.6 and 1, between 0.6 and 0.8, or between 0.8 and 1.

40. 40. The method of any one of claims 1 to 39, wherein in step (4), the patient is selected for IgG-RT when (s) is satisfied.

41. 41. The method of any one of claims 1 to 40, wherein in (s), Ts is 4, 3.5, 3, 2.5, or 2.

42. 41. The method of any one of claims 1 to 40, wherein in (s), Ts is between 2 and 4, between 2 and 3.5, between 2 and 3, between 2 and 2.5, between 2.5 and 4, between 2.5 and 3.5, between 2.5 and 3, between 3 and 4, between 3 and 3.5, or between 3.5 and 4.

43. The method of any one of claims 1 to 42, wherein in step (3), one, two, three, four, five, or six analyses of (a) to (f) are performed for the analysis of the BCR repertoire.

44. 43. The method of any one of claims 1 to 42, wherein in step (4), the patient is selected for IgG-RT when 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, or 13 criteria selected from (g) to (s) are met.

45. 43. The method of any one of claims 1 to 42, wherein in step (4), the patient is selected for IgG-RT when at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, or at least twelve criteria selected from (g) to (s) are met.

46. 46. ​​The method of any one of claims 1 to 45, wherein the patient is selected as having serum IgG less than 5 g / L and peripheral B cells greater than 40 / μL.

47. 46. ​​The method of any one of claims 1 to 45, wherein the patient is selected as having serum IgG less than 4.5 g / L and peripheral B cells greater than 35 / μL.

48. 46. ​​The method of any one of claims 1 to 45, wherein the patient is selected as having serum IgG less than 4 g / L and peripheral B cells greater than 30 / μL.

49. The method of any one of claims 1 to 48, wherein the patient sample comprises peripheral blood mononuclear cells (PBMCs).

50. 50. The method of any one of claims 1 to 49, further comprising sequencing said at least 10,000 transcripts, thereby providing said sequence information.

51. 51. The method of any one of claims 1 to 50, further comprising the step of treating said patient with IgG-RT if said patient is selected for IgG-RT.

52. 52. A method of treating a patient with hypogammaglobulinemia, comprising administering immunoglobulin replacement therapy (IgG-RT) to said patient, wherein said patient has been selected for IgG-RT using the method of any one of claims 1-51.

53. 52. A method of treating a hypogammaglobulinemic patient, further comprising selecting said patient for IgG-RT using the method of any one of claims 1 to 51.

54. 1. A diagnostic product for selecting hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT), wherein said diagnostic product is stored on a non-transitory computer readable medium and comprises: (1) obtaining sequence information for a plurality of patients with hypogammaglobulinemia, wherein the sequence information for each patient comprises the sequences of at least 10,000 transcripts from a sample of said patients comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) characterizing the B cell receptor (BCR) repertoire of each of said patients by identifying antibody clones based on said sequence information; (3) obtaining a training dataset comprising a plurality of training examples, each training example corresponding to the BCR repertoire of an individual patient; and (c) one or more characteristics associated with the BCR repertoire of the individual patient; and (d) diagnosing said hypogammaglobulinemic patient as to whether said individual patient is in need of IgG-RT, optionally said diagnosis being based on information related to serum IgG levels or susceptibility to infection. and (4) numerically coding the training examples in the training dataset, including numerically coding the one or more characteristics associated with the BCR repertoire of the individual patient and numerically coding the diagnosis of the hypogammaglobulinemic patient as to whether the individual patient requires IgG-RT; (5) For a diagnostic model comprising a neural network having a plurality of layers, each layer having a plurality of parameters, the layer comprising an input layer for receiving the one or more numerically encoded features of the BCR repertoire, and an output layer indicative of the likelihood of need for IgG-RT, during one or more iterations of a training process: (e) applying the parameters of the neural network to the set of training examples of the current iteration to generate an estimated likelihood for the set of training examples; (f) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (g) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (h) stopping the backpropagation after the loss function meets a criterion; and (6) storing the updated set of parameters for the layer of the diagnostic model in the computer-readable storage medium; 1. A diagnostic product manufactured by a process comprising:

55. In step (3)(a), the one or more characteristics associated with the BCR repertoire of the individual patient are selected from the group consisting of 1) to 7), 8) the number or abundance of individual IgG clones in the BCR repertoire; 9) an IgM antibody clonal diversity index value calculated by measuring the number and abundance of individual IgM antibody clones within the BCR repertoire; 10) the total frequency of the top 10, top 11, top 12, top 13, top 14, top 15, top 16, top 17, top 18, top 19, top 20, top 21, top 22, top 23, top 24, top 25, top 26, top 27, top 28, top 29, or top 30 IgG antibody clones that are most frequently found in the BCR repertoire; 11) the total frequency of the top 10, top 11, top 12, top 13, top 14, top 15, top 16, top 17, top 18, top 19, top 20, top 21, top 22, top 23, top 24, top 25, top 26, top 27, top 28, top 29, or top 30 IgM antibody clones that are most frequently found in the BCR repertoire; 12) variable region (V) gene usage in the antibody clone, optionally wherein the V genes are selected from the group consisting of IGHV4-30-2 heavy chain V genes, IGHV4-30-4 heavy chain V genes, IGHV3-23 heavy chain V genes, IGHV4-34 heavy chain V genes, and IGHV4-31 heavy chain V genes; 13) the average percent germline identity measured by comparing the variable (V) or joining (J) regions in the BCR repertoire to the corresponding germline sequences; and 14) The median somatic nucleotide mutation frequency in the V region in the BCR repertoire, optionally the V region comprising FR1, CDR1, FR2, CDR2 and / or FR3 regions.

55. The diagnostic product of claim 54, selected from:

56. 1. A diagnostic product for selecting hypogammaglobulinemic patients for treatment with immune globulin replacement therapy (IgG-RT), wherein said diagnostic product is stored on a non-transitory computer readable medium and comprises: (1) obtaining sequence information for a plurality of patients with hypogammaglobulinemia, wherein the sequence information for each patient comprises the sequences of at least 10,000 transcripts from a sample of said patients comprising B cells, each of said transcripts encoding an IgG or IgM heavy chain or a portion thereof; (2) obtaining a training dataset comprising a plurality of training examples, wherein each training example comprises sequence information of said individual patient and a diagnosis of said individual patient as to whether said individual patient is in need of IgG-RT, optionally wherein said diagnosis is based on information related to serum IgG levels or susceptibility to infection; (3) numerically encoding the training examples in the training dataset, including numerically encoding the sequence information of the individual patient and numerically encoding the individual patient's diagnosis of whether the individual patient is in need of IgG-RT; (4) For a diagnostic model including a neural network having a plurality of layers, each layer having a plurality of parameters, the layer including an input layer for receiving the numerically coded sequence information and an output layer indicating the likelihood of need for IgG-RT, during one or more iterations of a training process: (e) applying the parameters of the neural network to the set of training examples of the current iteration to generate an estimated likelihood for the set of training examples; (f) calculating a loss function indicative of the discrepancy between the estimated likelihood and the numerically coded diagnosis for the set of training examples for the current iteration; (g) iteratively backpropagating one or more error terms obtained from the loss function to update the parameters of the layer of the diagnostic model; and (h) stopping the backpropagation after the loss function meets a criterion; and (5) storing the updated set of parameters for the layer of the diagnostic model in the computer-readable storage medium; 1. A diagnostic product manufactured by a process comprising:

57. 1. A method for selecting a hypogammaglobulinemic patient for immune globulin replacement therapy (IgG-RT), comprising: (1) obtaining sequence information for at least 10,000 transcripts from the patient sample containing B cells, wherein each of the transcripts encodes an IgG or IgM heavy chain or a portion thereof; (2) providing said sequence information or information related to the patient's B-cell receptor (BCR) repertoire obtained by processing said sequence information to a diagnostic product according to any one of claims 54 to 56 and operating said diagnostic product; (3) obtaining information from the diagnostic product regarding whether the patient requires IgG-RT; A method comprising:

58. The step (2) of characterizing the B cell receptor (BCR) repertoire comprises the steps of: (a) to (f): (g) determining the number or abundance of individual antibody clones in the BCR repertoire; (h) calculating a diversity index value for said antibody clones by measuring the number and abundance of individual antibody clones in said BCR repertoire; (i) selecting at least 10 but not more than 30 antibody clones that are most frequent in the BCR repertoire and calculating the total frequency of the 10 to 30 most frequent antibody clones; (j) determining variable region gene (V region) usage in said antibody clone, optionally wherein said V is selected from the group consisting of an IGHV4-30-2 heavy chain V region gene, an IGHV4-30-4 heavy chain V region gene, an IGHV3-23 heavy chain V region gene, an IGHV4-34 heavy chain V region gene, and an IGHV4-31 heavy chain V region gene; (k) determining percent germline identity by comparing the V or J regions in said BCR repertoire with corresponding germline sequences; and (l) determining somatic mutation frequencies at different regions along the V region; 58. The method of claim 57, comprising analyzing the BCR repertoire by one or more steps selected from:

59. 59. The method of claim 57 or 58, further comprising treating the patient with IgG-RT if the patient is selected for IgG-RT.

60. 60. A method of treating a patient with hypogammaglobulinemia, comprising administering immunoglobulin replacement therapy (IgG-RT) to the patient, wherein the patient has been selected for IgG-RT using the method of claim 57 or 58.

Images