Biomarkers associated with drug induced interstitial lung disease

Abstract

There is disclosed a method for predicting the propensity for, and / or monitoring for, interstitial lung disease (ILD) in a subject receiving or beginning treatment for a cancer with an anti-cancer agent. The method comprises determining the amount of a baseline biomarker in the subject, wherein the baseline biomarker is one or more biomarkers selected from the group consisting of ST2 (IL1RL1), PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1. There is also disclosed a method of predicting an elevated propensity for ILD or an onset of ILD using longitudinal biomarkers, and a method of screening a subject for treatment with an antibody-drug conjugate in which the drug-linker comprises deruxtecan.

Description

[0001] BIOMARKERS ASSOCIATED WITH DRUG INDUCED INTERSTITIAL LUNG DISEASE

[0002] Related Applications

[0003] This application claims priority to and the benefit of United States Provisional Application No. 63 / 776,884, filed March 24, 2025, which is incorporated herein by reference in its entirety.

[0004] Reference to Sequence Listing

[0005] The content of the electronically submitted sequence listing “PN855069WO ST26 seq listing XML final for filing”, size 13.6KB, created 17 March 2026, submitted concurrently with this application is incorporated by reference in its entirety.

[0006] Technical Field

[0007] The present invention relates to methods of predicting the propensity for, and monitoring for, interstitial lung disease, and to methods of screening subjects for treatment.

[0008] Background

[0009] Trastuzumab deruxtecan (T-DXd) is a HER2-directed antibody-drug conjugate (ADC) in which the drug-linker is deruxtecan that includes a cleavable tetrapeptide-based linker. Deruxtecan (DXd) is a DNA topoisomerase I inhibitor and cytotoxic anticancer agent. T-DXd has demonstrated strong efficacy in multiple cancer indications, and has been approved for several indications (Keam SJ, Drugs, 2020 April; 8(5): 501-508). However, T-DXd treatment has been associated with interstitial lung disease (ILD) in 15.4% of patients across nine monotherapy trials (Powell et al., ESMO Open 2022, doi.org / 10.1016 / j.esmoop.2022.100554).

[0010] ILD encompasses a group of lung diseases characterised by inflammation and fibrosis of the lungs, resulting in breathing difficulties. The damage to the lungs is usually irreversible, and usually worsens with time. The diagnosis of drug-induced ILD (DI-ILD) is challenging and requires a combination of physiological, clinical, radiological and sometimes histopathological assessment which most often provides heterogeneous and non-specific information. Currently, ILD / pneumonitis is empirically determined by new or worsening pulmonary symptoms (e.g., dyspnoea, cough, or fever) or radiological abnormality suggestive of ILD / pneumonitis. Evaluation for ILD can include physical signs and symptoms (cough, shortness of breath, and pyrexia, etc); patient’s detailed past medical history, including concomitant medications; auscultation of lung field; arterial blood gases;

[0011] 15439666-1pulmonary function tests and pulse oximetry (SpO2); blood test; bronchoscopy and bronchoalveolar lavage; and HRCT. If a competing etiology is confirmed based on these evaluations, then drug-induced ILD can be ruled out. If another etiology cannot be identified, then a patient may be classified as having drug-induced ILD. The grade of ILD ranges from 1 to 5, with grades 1-2 being considered mild / moderate and grades 3-5 being considered severe ILD (see, e.g., National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE)). When a patient potentially has ILD, the anti-cancer treatment regimen is temporarily interrupted, to allow a determination of whether the ILD is drug induced. Diagnosis delays are therefore common, which can lead to preventable disease progression or even death. Thus, there is a need for early and accurate diagnosis of DI-ILD.

[0012] In addition, DI-ILD is difficult to manage - the disease can progress differently between patients, and progress is difficult to predict. Treatment most often relies on therapy interruption and / or steroid use, which has limited efficacy and doesn’t benefit all the patients equally. Hence, a greater understanding of DI-ILD pathophysiology is also needed to provide more effective therapeutic options for patients that will develop ILD.

[0013] It has been found that ADCs with the same drug-linker but different antibodies can exhibit similar toxicities (Mahalingaiah PK, et al., Pharmacol Ther. 2019:200:110-125). For example, ocular (corneal) toxicity has reported as a dose limiting toxicity (DLT) for multiple DM4-containing ADCs targeting unrelated antigens not expressed in the eye (de Goeij & Lambert, Current Opinion in Immunology, 2016, 40: 14-23).

[0014] T-DXd has been reported to induce interstitial pneumonitis in cynomolgus monkeys similar to drug-induced ILD in humans (Kumagai K, et al., Cancer Sci. 2020;111(12):4636-4645).

[0015] The detailed pathogenesis of T-DXd-related ILD remains unclear. As with other drug-induced ILD, direct cytotoxicity by the cytotoxic payload and / or an immune-mediated mechanism of action is speculated (Swain SM, et al., Cancer Treat Rev.

[0016] 2022:106:102378; Abuhelwa Z, et al., Drugs. 2022;82(9):979-987; Wekking D, et al., ESMO Open. 2023; 8(6): 102043). For T-DXd-related ILD, an exposure-response relationship with steady-state AUC of plasma T-DXd for all grades and steady-state Cmax of plasma T-DXd for grades 3 and above have been suggested (Yin O, et al., Clin Pharmacol Ther. 2021 ;110(4): 986-996). Powell et al., supra, also demonstrated through nine T-DXd monotherapy trials that clinical features of age, oxygen saturation, renal

[0017] 15439666-1impairment, lung comorbidities, and enrolment in Japan were associated with increased risk of T-DXd drug related adjudicated ILD.

[0018] Thus, there is a need to be able to predict and monitor the risk of developing ILD, and particularly severe ILD, as well as a need to improve the prevention and diagnosis of ILD.

[0019] Summary of Invention

[0020] The present disclosure arises from the observation from various studies of cancer patients undergoing treatment with trastuzumab deruxtecan (T-DXd) that differences in the amounts of certain proteins in serum samples from the patients prior to treatment initiation are associated with an increased risk of onset of DI-ILD, and particularly severe ILD. These proteins are therefore useful as biomarkers for predicting and monitoring the likelihood of development of ILD. Moreover, changes in the amounts of some of these proteins are associated with an increased risk of developing ILD, particularly severe ILD, during the ADC treatment course, before the onset of ILD, such that these proteins can be used to longitudinally monitor the change in the level of risk of a subject of developing ILD. Thus, instead of diagnosing ILD by eliminating alternative diseases, the biomarkers disclosed herein can inform of a patient’s risk of developing DI-ILD or status of DI-ILD before treatment, during treatment, prior to onset of ILD, during onset of ILD, and after mitigation of the ILD.

[0021] Thus, in in a first aspect, there is provided a method of predicting the propensity for, and / or monitoring for, interstitial lung disease (ILD) in a subject receiving or beginning treatment for a cancer with an anti-cancer agent, the method comprising: determining the amount of a baseline biomarker in the subject, wherein the baseline biomarker is one or more biomarkers selected from the group consisting of ST2 (IL1RL1), PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1.

[0022] Preferably, the baseline biomarker is selected from ST2 (IL1RL1) and PASP (CPB1).

[0023] Conveniently:

[0024] i) an increase in ST2 (IL1RL1);

[0025] ii) a decrease in the amount of one or more biomarker selected from the group consisting of PASP (CPB1), FGFBP1, SCUBE3, SFRP1, and NTN1;

[0026] iii) an increase in the ratio of ST2 (IL1RL1):PASP (CPB1); and / or

[0027] iv) an increase in the ratio of ST2 (IL1RL1) to any one of FGFBP1, SCUBE3, SFRP1, and NTN1;

[0028] 15439666-1is indicative of an elevated propensity for ILD in the subject.

[0029] Preferably, an increase or decrease in the amount of the baseline biomarker or an increase in the ratio is with respect to an amount of the baseline biomarker or the ratio previously determined in the subject, with respect to an amount of the baseline biomarker or the ratio in an individual not having ILD, or with respect to an index of the baseline biomarker or the ratio calculated from a plurality of individuals not having ILD.

[0030] Advantageously, the method further comprises, based on the elevated propensity for ILD, adjusting a treatment regimen for the subject by one or more of the following:

[0031] i) administering a reduced amount of the anti-cancer agent, the reduced amount being less than a therapeutically effective amount of the anti-cancer agent that would have been prescribed to the subject absent the elevated propensity for ILD;

[0032] ii) discontinuing treatment with the anti-cancer agent;

[0033] iii) delaying treatment with the anti-cancer agent until the subject’s baseline biomarker or the ratio of baseline biomarkers is no longer increased;

[0034] iv) prophylactically treating the subject for the ILD with a steroid or other prophylactic therapeutic intervention for ILD; and / or

[0035] v) monitoring the subject for progression or development of ILD.

[0036] Conveniently, the monitoring further comprises predicting an elevated propensity for ILD or an onset of ILD in the subject if one or more of:

[0037] i) the amount of one or more longitudinal biomarkers selected from SLAMF7, SAA1, SAA2, CRP, IL-5RA, CXCL9, CXCL10, CXCL11, CXCL13, GBP1, SP-D (SFTPD), IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), PD-L1 (CD274), SFN, IL-9, HSP-90 (HSP90AB1), IL-18BP, PAPPA, FLRT3, CD87 (PLAUR), TNFAIP6, RELN, AFM, LDH-H (LDHB), ICAM1, RBP4, IL-6, AKR1A1, ITIH3, IDO1, RPS20, ILT-6 (LILRA3), CLIC3, and STAT 1 is increased at a second time point compared to a first time point;

[0038] ii) the amount of one or more longitudinal biomarkers is selected from SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI- 2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, IL-9, IL-6, HSP-90 (HSP90AB1), TNFAIP6, RELN, AFM and RBP4 is increased at the second time point compared to the first time point;

[0039] iii) the amount of TNFAIP6 is increased at the first time point and the amount of one or more biomarker selected from the group consisting of CXCL11, PAI-2 (SERPINB2), CAP- 3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53

[0040] 15439666-1(WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point;

[0041] iv) the amount of CXCL11 is increased at the first time point and the amount of one or more biological marker selected from the group consisting of PAI-2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point;

[0042] v) the amount of CXC11 is increased at the second time point compared to the first time point; and / or

[0043] vi) the amount of one or more longitudinal biomarkers selected from the group consisting of TNFAIP6, CXCL11, CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1) and IL-18BP is increased at the first time point and the amount of one or more longitudinal biomarkers selected from the group consisting of SP-D, PAPP, CXCL9, CXCL10, PAI-2 (SERPINB2), AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point,

[0044] wherein, in i)-vi), the increase at the second time point may be with respect to an amount determined at the first time point or another previously determined amount of the biomarker in the subject, with respect to an amount of the biomarker in an individual not having ILD, or with respect to an index of the biomarker calculated from a plurality of individuals not having ILD.

[0045] Preferably, if the subject is predicted to have an elevated propensity for ILD, the method further comprises adjusting a treatment regimen for the subject by one or more of the following:

[0046] i) administering a reduced amount of the anti-cancer agent, the reduced amount being less than a therapeutically effective amount of the anti-cancer agent that would have been prescribed to the subject absent the elevated propensity for ILD;

[0047] ii) discontinuing treatment with the anti-cancer agent;

[0048] iii) delaying treatment with the anti-cancer agent until the subject’s one or more longitudinal biomarkers are no longer increased;

[0049] iv) prophylactically treating the subject for the ILD with a steroid or other prophylactic therapeutic intervention for ILD; and / or

[0050] v) monitoring the subject for progression or development of ILD.

[0051] Conveniently, the baseline biomarker is ST2 (IL1RL1) or is the ratio of ST2 (IL1RL1):PASP (CPB1).

[0052] 15439666-1Advantageously, the longitudinal biomarker is SP-D, CXCL10, and / or CXCL11.

[0053] Advantageously:

[0054] i) if the subject’s propensity for ILD is determined to be increased, then the subject is administered with a treatment for ILD, and / or the dosage of an ongoing treatment of the subject is reduced or an ongoing treatment of the subject is discontinued, or

[0055] ii) if the subject’s propensity for ILD is determined to be not increased,

[0056] then the subject is administered with a treatment for which the subject is in need, or the dosage of an ongoing treatment of the subject is maintained or increased.

[0057] Preferably, the method is performed prior to onset of ILD in the subject.

[0058] Advantageously, the ILD is severe ILD.

[0059] Conveniently, the ILD is drug induced-ILD.

[0060] Preferably, the amount of the baseline biomarker is in a biological sample obtained from the subject; and / or each of the biomarker obtained at the first time point and the biomarker obtained at the second time point is in a respective biological sample obtained from the subject.

[0061] Conveniently, the biological sample is a plasma sample, a bronchoalveolar lavage (BAL) fluid sample, a sputum sample, a blood sample, or a serum sample.

[0062] Advantageously, the biological sample is a serum sample.

[0063] Conveniently, the cancer is at least one selected from the group consisting of breast cancer, lung cancer, colorectal cancer, gastric cancer, esophageal cancer, head-and-neck cancer, esophagogastric junction adenocarcinoma, biliary tract cancer, Paget's disease, pancreatic cancer, ovarian cancer, uterine carcinosarcoma, urothelial cancer, prostate cancer, bladder cancer, gastrointestinal stromal tumor, digestive tract stromal tumor, uterine cervix cancer, squamous cell carcinoma, peritoneal cancer, liver cancer, hepatocellular cancer, corpus uteri carcinoma, kidney cancer, vulval cancer, thyroid cancer, penis cancer, leukemia, malignant lymphoma, plasmacytoma, myeloma, glioblastoma multiforme, osteosarcoma, sarcoma, melanoma, cervical cancer, uterine cancer, testicular cancer, and renal cell carcinoma.

[0064] 15439666-1Preferably, the cancer is at least one selected form the group consisting of breast cancer, lung cancer, gastric cancer, colorectal cancer or non-small cell lung cancer.

[0065] Advantageously, the anti-cancer agent, an ongoing anti-cancer treatment, or a treatment for which the subject is in need is treatment with an antibody-drug conjugate in which a drug-linker represented by the following formula:

[0066]

[0067] wherein A represents the connecting position to an antibody via a thioether bond.

[0068] Conveniently, the antibody is an anti-HER2 antibody.

[0069] Preferably, the antibody-drug conjugate is trastuzumab deruxtecan (DS-8201).

[0070] In a second aspect, there is provided a method of screening a subject for a treatment with an antibody-drug conjugate in which a drug-linker is represented by the following formula:

[0071]

[0072] wherein A represents the connecting position to an antibody via a thioether bond, the method comprising:

[0073] determining the amount of a biomarker in a subject and comparing the amount to an index of the biomarker, wherein if the determined amount of the biomarker relative to

[0074] 15439666-1the index is indicative of a propensity for ILD, then the subject is not selected for the treatment, and if the determined amount of the biomarker relative to the index is not indicative of a propensity for ILD, then the subject is selected for the treatment, and wherein the biomarker is one or more baseline biomarkers selected from the group consisting of ST2 (IL1RL1), PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1.

[0075] Preferably, the subject is selected for the treatment when:

[0076] i) the amount of the baseline biomarker ST2 (IL1RL1) is the same as or below the index of the biomarker;

[0077] ii) the baseline biomarker is selected from a group consisting of PASP (CPB1), FGFBP1, SCLIBE3 and SFRP1 , and the amount of the biomarker is the same as or above the index of the biomarker, and / or

[0078] iii) the biomarker is the ratio of ST2 (IL1RL1):PASP (CPB1), and the ratio is the same as or below the index of the biomarker,

[0079] wherein the index of the biomarker is calculated from a plurality of individuals not having ILD.

[0080] Conveniently, if the subject is selected for the treatment, the treatment is administered to the subject.

[0081] Advantageously, the antibody is an anti-HER2 antibody.

[0082] Preferably, the antibody-drug conjugate is trastuzumab deruxtecan (T-DXd).

[0083] Conveniently, the anti-HER2 antibody is an antibody comprising a heavy chain comprising CDRH1 consisting of an amino acid sequence represented by SEQ ID NO: 3, CDRH2 consisting of an amino acid sequence represented by SEQ ID NO: 4 and CDRH3 consisting of an amino acid sequence represented by SEQ ID NO: 5, and a light chain comprising CDRL1 consisting of an amino acid sequence represented by SEQ ID NO: 6, CDRL2 consisting of an amino acid sequence consisting of amino acid residues 1 to 3 of SEQ ID NO: 7 and CDRL3 consisting of an amino acid sequence represented by SEQ ID NO: 8.

[0084] Advantageously, the anti-HER2 antibody is an antibody comprising a heavy chain comprising a heavy chain variable region consisting of an amino acid sequence represented by SEQ ID NO: 9 and a light chain comprising a light chain variable region consisting of an amino acid sequence represented by SEQ ID NO: 10.

[0085] 15439666-1Preferably, the anti-HER2 antibody is an antibody comprising a heavy chain consisting of an amino acid sequence represented by SEQ ID NO: 1 and a light chain consisting of an amino acid sequence represented by SEQ ID NO: 2.

[0086] Preferably, the anti-HER2 antibody is an antibody comprising a heavy chain consisting of an amino acid sequence represented by SEQ ID NO: 11 and a light chain consisting of an amino acid sequence represented by SEQ ID NO: 2.

[0087] In a third aspect, there is provided a method of predicting severe ILD in a subject receiving treatment for a cancer, wherein the treatment comprises a dosage amount of trastuzumab deruxtecan, the method comprising:

[0088] determining that the subject has an increased amount of a baseline biomarker comprising ST2 (IL1RL1) or a ratio of ST2 (IL1RL1):PASP (CPB1), indicating that the subject has an elevated risk for developing severe ILD;

[0089] based on the elevated risk for developing severe ILD, measuring samples collected from the subject during treatment with trastuzumab deruxtecan for one or more longitudinal biomarker for ILD selected from CXC10, CXC11, and SP-D;

[0090] adjusting the subject’s treatment regimen of trastuzumab deruxtecan if a later collected sample from the subject contains an elevated amount of the one or more longitudinal biomarker compared to an earlier collected sample by one or more of the following ways: i) reducing the dosage amount and / or frequency of trastuzumab deruxtecan, the reduced amount being less than a therapeutically effective amount of the anti-cancer agent that would have been prescribed to the subject absent the elevated propensity for ILD; ii) discontinuing treatment with trastuzumab deruxtecan;

[0091] iii) discontinuing or delaying treatment with trastuzumab deruxtecan until a subsequently collected sample from the subject exhibits a decrease or comparable amount of the longitudinal biomarker relative the later or the earlier sample of the subject, the subsequently collected sample being collected after the later sample; and / or

[0092] iv) prophylactically treating the subject for the ILD with a steroid or other prophylactic therapeutic intervention for ILD.

[0093] In a fourth aspect, there is provided a method of predicting the propensity for, and / or monitoring for, interstitial lung disease (ILD) in a subject receiving or beginning treatment for a cancer with an anti-cancer agent, the method comprising:

[0094] determining the amount of a baseline biomarker in the subject, wherein the biomarker is ST2 (IL1RL1),

[0095] 15439666-1wherein an increase in ST2 (IL1RL1) is indicative of an elevated propensity for ILD in the subject,

[0096] wherein the increase in the amount of the baseline biomarker is with respect to an amount of the baseline biomarker previously determined in the subject, with respect to an amount of the baseline biomarker in an individual not having ILD, or with respect to an index of the baseline biomarker calculated from a plurality of individuals not having ILD, wherein i) if the subject’s propensity for ILD is determined to be increased, then the subject is administered with a treatment for ILD, and / or the dosage of an ongoing treatment of the subject is reduced or an ongoing treatment of the subject is discontinued, or

[0097] ii) if the subject’s propensity for ILD is determined to be not increased, then the subject is administered with a treatment for which the subject is in need, or the dosage of an ongoing treatment of the subject is maintained or increased, wherein the ILD is severe drug induced-ILD, wherein the subject has cancer, wherein the cancer is at least one selected form the group consisting of breast cancer, lung cancer, gastric cancer, colorectal cancer and non-small cell lung cancer, and wherein the ongoing treatment or the treatment for which the subject is in need is treatment with trastuzumab deruxtecan (T-DXd).

[0098] As used herein, the term “propensity for ILD” means the subject’s likelihood or risk of developing ILD. An increased propensity means that the subject has a higher than average likelihood / risk of developing ILD.

[0099] Brief Description of the Figures

[0100] Figure 1(A) shows the design of the study used to obtain the data reported herein, as well as sampling and cohort information; Figure 1(B) shows the worst ILD grade of subjects in each clinical study (DESTINY Breast 01 (DB-01), DESTINY Breast 04 (DB-04) and DESTINY Lung 01 (DL-01)). For each of the groups listed (e.g., No ILD, Gr 1, Gr2, GR 3, Gr 5), the columns from left to right correspond to DB-01, DB-04 and DL-01; and Figure 1(C) the time to worst ILD diagnosis for patients in each clinical study.

[0101] Figure 2(A) shows a volcano plot of the results of differential analysis of proteins in patients who developed severe ILD in comparison to patients who did not develop any grade ILD (SomaLogic); Figure 2(B) shows box plots of the abundance of ST2 (IL1RL1) protein in each of the three clinical studies (DB-01, DB-04 and DL-01) and after combination of all three cohorts, according to ILD grading (no ILD (left plot / box), mild / moderate ILD (grade 1 or 2, middle plot / box) and severe ILD (grade 3 or above, right plot / box), respectively, for each), using both Somalogic® assay (top row) and Luminex® assay (bottom row); and Figure 2(C) shows the results of optimal cut-point analysis of ST2

[0102] 15439666-1protein using the Youden index approach. For each group shown in Figure 2B (e.g., All, DB-01, DB-04, DL-01), the box plots left to right correspond to no ILD, mild / moderate grade ILD (Grade 1 or 2), and severe ILD (Grade 3+); Figure 2(D) shows that the speed of ILD onset is correlated with the baseline concentration of ST2 for patients who develop severe, but not mild ILD; Figure 2(E)i) shows that patients who develop severe ILD have heightened neutrophil levels at baseline. For each box plot shown for all clinical studies or DB-01 , DB-04, or DL-01 , the box plots left to right correspond to no ILD, mild / moderate grade ILD (Grade 1 or 2), and severe ILD (Grade 3+). Figure 2(E)ii) shows that neutrophil counts are correlated with ST2 at baseline and correlation is greatest in patients who develop drug induced I LD. Figure 2(F) shows the cut point analysis using neutrophil count at baseline. In addition, correlation of ST2 with other proteins indicates a relationship between ST2 and neutrophil biology. Figure 2(G) shows longitudinal comparisons of ST2 (IL1RL1), ST2 (IL1RL1):PASP (CPB1) ratio and neutrophil counts for no ILD subjects, mild ILD subjects, and severe ILD subjects (left to right, respectively, for each group shown for each cycle / day count for each comparison).

[0103] Figure 3(A) shows a scatterplot of proteins which are downregulated in the severe ILD patient population; Figure 3(B) shows boxplots of the abundance of proteins found to be differentially downregulated in severe I LD patients, with each boxplot showing, left to right, the abundance of the respective protein in each of patients with no ILD, mild ILD and severe ILD, respectively; Figure 3(C) shows a scatterplot of a SomaScan® assay dataset; Figure 3(D) shows a sensitivity-specificity curve for each of ST2 (IL1RL1), PASP (CPB1) and the ratio of ST2 (IL1RL1):PASP (CPB1) (SomaLogic®). This graph shows that using both ST2 (IL1 RL1) and PASP (CPB1) improves severe ILD onset prediction in comparison to each protein individually. The data for ST2 (IL1RL1) is shown in the lightest gray text or lines, the data for PASP (CPB1) is shown in an intermediate gray, and the data for the ratio of ST2 (IL1RL1):PASP (CPB1) is shown in a dark gray; and Figure 3(E) shows boxplots showing the ratio of ST2 (IL1RL1):PASP (CPB1) in each of the three clinical studies (DB-01, DB-04 and DL-01) and after combination of all three cohorts, according to ILD grading (left to right: no ILD, mild ILD and severe ILD, respectively) for each of the clinical studies.

[0104] Figure 4(A) shows a scatterplot of proteins found to be significant in all-grade ILD and severe ILD patients, with shading indicating the Spearman coefficient of correlation; Figure 4(B) shows a cluster map of proteins associated with ILD onset grouped by biological mechanism; and Figure 4(C) shows boxplots of the abundance of specific proteins at the at worst ILD diagnosis for each of no ILD, mild ILD and severe ILD (left to

[0105] 15439666-1right in each boxplot), with the top row showing proteins associated with systemic inflammation, the second row showing proteins associated with local inflammation, the third row showing proteins associated with lung damage and the bottom row showing proteins associated with fibrosis.

[0106] Figure 5(A) shows the results of differential analysis (severe ILD vs no ILD) of proteins at 0-24 days, 25-49 days, 50-74 days and 75-99 days before worst ILD diagnosis; Figure 5(B) shows boxplots of the change in abundance of specific proteins at each of 100-124 days, 75-99 days, 50-74 days, 25-49 days and 0-24 days before worst ILD diagnosis in patients with no ILD, mild ILD and severe ILD (left to right, respectively, for each time window for each protein); Figure 5(C) shows boxplots show protein abundance as fold change relative to the study start, Cycle 1, Day 1 (C1D1) at each treatment cycle (from Cycle 2 Day 1 (C2D1) to Cycle 6 Day 1 (C6D1)) in patients having no ILD, mild ILD or severe ILD at ILD (left to right), respectively, for each treatment cycle). ILD is classification at onset is used, where ILD onset means first diagnosis. Patients develop ILD at different rates (see, e.g., FIG. 1B and 1C). Cycles denote treatment cycles where patients receive a dose (approximately q3w); Figure 5(D) shows sensitivity-selectivity curves resulting from optimal cut-point analysis of CXCL11 across specific time windows and treatment cycles; and Figure 5(E) shows scatterplots showing the results of SomaScan® assay-Olink® assay and SomaScan® assay- Luminex® assay correlations of specific proteins in DESTINY Lung 01 (DL-01) samples.

[0107] Figure 6(A) depicts boxplot results of ST2 levels as detected in pooled cohort analyses by Luminex® assay; Figure 5(B) and 5(C) show boxplot results for levels of ST2 as detected in individual cohort analyses by Luminex® assay as described in Example 6.

[0108] Figure 7 depicts results for detection of ST2 by Luminex® assay (7A) and Quantikine® assay (7B), as well as concordance among the assays (7C) as described in Example 7.

[0109] Figure 8(A-B) depict boxplot results for ST2 levels as detected in pooled studies analyses by Quantikine® assay as described in Example 8.

[0110] Figure 9(A-H) depict boxplot results for ST2 levels as detected in individual study analyses by Quantikine® assay as described in Example 8.

[0111] Figure 10(A-B) depict results of correlation studies of SpO2 and NLR and ST2 levels in samples as described in Example 8.

[0112] 15439666-1Brief Description of the Sequence Listings

[0113] SEQ ID NO: 1 is the amino acid sequence of heavy chain of the antibody trastuzumab. SEQ ID NO: 2 is the amino acid sequence of the light chain of the antibody trastuzumab. SEQ ID NO: 3 is the amino acid sequence of the heavy chain CDRH1 of an anti-HER2 antibody.

[0114] SEQ ID NO: 4 is the amino acid sequence of the heavy chain CDRH2 of an anti-HER2 antibody.

[0115] SEQ ID NO: 5 is the amino acid sequence of the heavy chain CDRH3 of an anti-HER2 antibody.

[0116] SEQ ID NO: 6 is the amino acid sequence of the light chain CDRL1 of an anti-HER2 antibody.

[0117] SEQ ID NO: 7 is the amino acid sequence of the light chain CDRL2 of an anti-HER2 antibody.

[0118] SEQ ID NO: 8 is the amino acid sequence of the light chain CDRL3 of an anti-HER2 antibody.

[0119] SEQ ID NO: 9 is the amino acid sequence of the heavy chain variable region of an anti-HER2 antibody.

[0120] SEQ ID NO: 10 is the amino acid sequence of the light chain variable region of an anti-HER2 antibody.

[0121] SEQ ID NO: 11 is the amino acid sequence of the heavy chain of an anti-HER2 antibody.

[0122] Detailed Description

[0123] In order to improve the understanding and pathophysiology, and the diagnosis and monitoring, of DI-ILD, a proteomics analysis was performed in 1200+ pre- and on-treatment serum samples from 190 patients (95 non-ILD, 73 mild ILD [grades 1-2] and 22 severe ILD [grades 3-5]) across three T-DXd clinical studies (DESTI NY-Breast01 [HER2-positive breast cancer, NCT03248492], DESTI NY-BreastO4 [HER2-low breast cancer, NCT03734029], DESTINY-LungOI [HER2-mutant(e.g., HER2 overexpressing or mutated) non-small-cell lung cancer, NCT03505710]) using two discovery proteomics platforms (i.e. SomaScan® assay and Olink® assay). Pre-treatment differential analysis and optimal cut point analysis identified dysregulated proteins in patients who developed severe ILD, indicative of subclinical pre-existing conditions in those patients. In particular, a protein marker of inflammation known to be associated with various pulmonary diseases (i.e. idiopathic pulmonary fibrosis and chronic obstructive pulmonary disease) could predict effectively severe ILD (AUG: 0.75). Importantly, this biomarker (ST2) performed well in each cohort separately, irrespective of patients’ demographics and cancer

[0124] 15439666-1indications. Longitudinal analysis was performed using linear mixed effect modelling and revealed several promising potential ILD diagnostic biomarkers (FDR < 0.05) that capture key facets of ILD pathophysiology, including alveolar damage, inflammation and fibrosis. Those biomarkers, as disclosed herein, may be useful for early detection of ILD and ultimately enable more efficient ILD management and reduction of T-DXd discontinuation rates.

[0125] Thus, in general terms, the present application concerns a method for predicting and / or monitoring a subject’s propensity for developing ILD, particularly severe ILD, by determining the amount of a biomarker in a subject, and optionally monitoring the amount of the biomarker in the subject over time. The subject may, for example, be undergoing or about to undergo treatment for a different disease or condition (i.e. not ILD), which may give rise to an increased propensity for ILD and / or which may be associated with onset of ILD. The present application also concerns a method for screening for subjects who do not have a propensity for ILD, or who have a low propensity for ILD, and who are therefore more suitable for receiving a treatment which is associated with onset of ILD. Conversely, the screening method also permits the identification of subjects who are not suitable for receiving a treatment which is associated with onset of ILD, because they are determined as having a propensity for ILD.

[0126] Moreover, the amount of the biomarker is useful for aiding the diagnosis of ILD, by indicating that there is an increased risk, and therefore an increased likelihood, of ILD. For example, when the amount of the biomarker indicates that the subject has a propensity for ILD, or that their propensity for ILD has increased, this information contributes to determining a diagnosis of the patient (e.g. it accelerates the diagnosis of ILD by indicating an increased probability that the disease / condition is ILD).

[0127] Biomarker

[0128] A baseline biomarker is one or more of the proteins ST2 (IL1RL1) (e.g., UniProtKB Q01638), PASP (CPB1) (e.g., UniProtKB P15086), NET1 (NTN1) (e.g., UniProtKB 095631), FGFBP1 (e.g., UniProtKB Q14512), SFRP1 (e.g., UniProtKB Q8N474), and / or SCUBE3 (e.g., UniProtKB Q8IX30). The baseline biomarker may be useful to predict a propensity for a subject to develop severe ILD before treatment with an anti-cancer therapy. Preferably, the baseline biomarker is ST2 (IL1RL1) or PASP (CPB1). Even more preferably, it is a combination of ST2 (IL1RL1) and PASP (CPB1). For example a ratio of ST2 (IL1RL1):PASP (CPB1), as disclosed herein, is useful to assess risk of developing severe ILD. A ratio of ST2(IL1RL1) to other baseline biomarkers may also be useful.

[0129] 15439666-1A longitudinal biomarker is one or more of the proteins AFAM (AFM) (e.g., UniProtKB P43652), SAA1 (e.g., UniProtKB P0DJI8), CXCL9 (e.g., UniProtKB Q07325), CXCL10 (e.g., UniProtKB P02778), CXCL11 (e.g., UniProtKB 014625), CXCL13 (e.g., UniProtKB 043927), SLAMF7 (e.g., UniProtKB Q9NQ25), SAA2 (e.g., UniProtKB P0DJI9), CRP (e.g., UniProtKB P02741), IL-5RA (e.g., UniProtKB Q01344), GBP1 (e.g., UniProtKB P32455), SP-D (SFTPD) (e.g., UniProtKB P35247), IFP53 (WARS1) (e.g., UniProtKB P23381), PAI-2 (SERPINB2) (e.g., UniProtKB P05120), CAP-3 (SERPINB9) (e.g., UniProtKB P50453), PD-L1 (CD274) (e.g., UniProtKB Q9NZQ7), SFN (e.g., UniProtKB P31947), IL-9 (e.g., UniProtKB Q01113), HSP-90 (HSP90AB1) (e.g., UniProtKB P08238), IL-18BP (e.g., UniProtKB 095998), PAPPA (e.g., UniProtKB Q13219), FLRT3 (e.g., UniProtKB Q9NZU0), CD87 (PLAUR) (e.g., UniProtKB Q03405), TNFAIP6 (e.g., UniProtKB P98066), RELN (e.g., UniProtKB P78509), RBP4 (e.g., UniProtKB P02753), LDH-H (LDHB) (e.g., UniProtKB P07195), ICAM1 (e.g., UniProtKB P05362), IL-6 (e.g., UniProtKB P05231), AKR1A1 (e.g., UniProtKB P14550), ITIH3 (e.g., UniProtKB Q06033), ID01 (e.g., UniProtKB P14902), RPS20 (e.g., UniProtKB P60866), ILT-6 (LILRA3) (e.g., UniProtKB Q8N6C8), CLIC3 (e.g., UniProtKB 095833), and STAT1 (e.g., UniProtKB P42224). Preferably, the longitudinal biomarker is one or more of SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDHB, ICAM1, PD-L1, IL-18BP, IL-9, IL-6, HSP-90 (HSP90AB1), TNFAIP6, RELN, AFM and RBP4. A longitudinal biomarker may be useful to monitor risk of the subject to develop ILD during treatment of the subject with an anti-cancer agent or therapy (e.g., T-DXd).

[0130] Determining the amount of a biomarker in a subject

[0131] In some embodiments, an increase in the amount of one or more of the following biomarkers is indicative of a propensity for or the presence of ILD in the subject: ST2 (IL1RL1), SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, GBP1, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, RBP4, IL-9, IL-6, HSP-90 (HSP90AB1), PAPPA, FLRT3, CD87 (PLAUR), AKR1A1, ITIH3, ID01, TNFAIP6, RPS20, ILT-6 (LILRA3), CLIC3 and STAT1. In these embodiments, the biomarker is preferably one or more of ST2 (IL1RL1), SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, RBP4, IL-9, IL-6, HSP-90 (HSP90AB1), and TNFAIP6. More preferably, the biomarker is one or more of ST2 (IL1RL1), SLAMF7, SAA1, SAA2, CRP, CXCL9, CXCL10, CXCL11, CXCL13, IL-5RA, SP-D, SFN, PD-L1, IL-18BP, HSP-90 (HSP90AB1),

[0132] 15439666-1PAI-2 (SERPINB2), CAP-3 (SERPINB9), IL-6, STAT1, IL-18BP and TNFAIP6. In particular, ST2 (IL1RL1) was identified as the most significant elevated protein in subjects who developed severe ILD compared to those who did not develop ILD (Figure 2A), and longitudinal analysis showed that increased amounts of, in particular, SLAMF7, SAA1, SAA2, CRP, CXCL9, CXCL10, CXCL11, CXCL13, IL-5RA, SP-D, SFN, PD-L1, IL-18BP, HSP-90 (HSP90AB1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), IL-6, STAT1, IL-18BP and TNFAIP6 are associated with onset of severe ILD (Figures 4C and 5A). Conversely, if there is a decrease or an absence of an increase of the amount of the one or more biomarker, it may determined that there is no indication of a propensity for ILD, or that the subject has a low propensity for ILD, particularly severe ILD.

[0133] In some embodiments, a decrease in the amount of one or more of the following baseline biomarkers may be indicative of a propensity for or the presence of ILD in the subject: PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1. In these embodiments, the biomarker is preferably PASP (CPB1). In particular, PASP (CPB1) has been identified as the most relevant of these proteins for improving the prediction of severe ILD (Figure 3C). FGFBP1, SCUBE3 and SFRP1 may have similar results. Preferably, the decrease in the amount of one or more of PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1 is observed in combination with an increase in the amount of ST2 (IL1RL1). Conversely, if there is an increase or an absence of a decrease of the amount of the one or more baseline biomarker (PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1), particularly when observed with a decrease in the amount of ST2 (IL1RL1), it may be determined that there is no indication of a propensity for ILD, or that the subject has a low propensity for ILD.

[0134] In some embodiments, the baseline biomarker is ST2 (IL1RL1) and / or PASP (CPB1), and an increase in the ratio of ST2:PASP (CPB1) is indicative of a propensity for or the presence of ILD in the subject before treatment with an anti-cancer therapy (e.g., T-DXd). In particular, determination of the amount of PASP (CPB1) in combination with the determination of the amount of ST2 has been found to improve the prediction of the propensity for severe ILD compared to the use of the amount of ST2 (IL1 RL1) alone, with the positive predictive value (PPV) increasing by about 27% (Figure 3D). Conversely, if there is a decrease or an absence of an increase in the ratio, it may be determined that there is no indication of a propensity for ILD, or that the subject has a low propensity for ILD. In an embodiment, the baseline biomarker is ST2 (IL1RL1) and one or more of FGFBP1, SCUBE3, SFRP1 and / or NTN1. An increase in the ratio of ST2 to one or more of FGFBP1 , SCUBE3, SFRP1 , and / or NTN1 may be indicative of a propensity for or the

[0135] 15439666-1presence of ILD, particularly severe ILD, in the subject before treatment with an anticancer therapy.

[0136] The one or more baseline biomarkers or ratio thereof may be useful, as disclosed herein, to identify a subject who, before treatment with an anti-cancer therapy, may have a higher propensity to develop severe ILD. If a subject is classified as having a higher propensity to develop severe ILD, then the subject may be more closely monitored for ILD such as by measuring an amount of one or more longitudinal biomarkers (e.g., a risk monitoring biomarker). Similarly, the subject may receive a reduced amount of the anti-cancer therapy or discontinue the anti-cancer therapy. In some instances, the subject may be prophylactically treated for severe ILD with steroids or other prophylactic intervention.

[0137] The amount of the biomarker in a subject or an individual may be an absolute amount, including an absolute concentration, or a relative amount, including a relative concentration. When it is a relative amount, the amount may be relative to an amount of a reference biomarker in the subject or individual. A relative abundance (e.g., from SomaLogic RFU or Olink NXP) indicates that no standard curve is used and it may be subject to batch-to-batch variation.

[0138] The increase or decrease in the amount of the biomarker, or the increase in the ratio of ST2:PASP (CPB1), may be with respect to an amount of the biomarker or a ratio previously determined from the subject, with respect to an amount of the biomarker or a ratio in an individual not having ILD, or with respect to an index of the biomarker or ratio calculated from a plurality of individuals not having ILD. For example, the index of the biomarker or ratio may be a reference amount of the biomarker or the ratio derived from, or calculated on the basis of, an average amount of the biomarker or an average ratio calculated from a plurality of individuals not having ILD. When the biomarker in the subject is more than one biomarker (i.e. a panel of biomarkers), then the index may be a score based on a composite of a value for the amount of each biomarker in the panel in an individual or a plurality of individuals not having ILD, and each value may be weighted differently for the corresponding biomarker.

[0139] The presence and amount of the biomarker can be determined using any technique known in the art. For example, a proteomic assay, such as an immunoassay, ELISA, or liquid chromatography - mass spectrometry (LC-MS), may be used to determine an amount of one or more biomarkers.

[0140] 15439666-1Preferably, the following changes indicate that a subject’s propensity for severe ILD has increased:

[0141] i) The subject has an elevated baseline amount of ST2 (IL1 RL1 ) ;

[0142] ii) The subject has an elevated baseline amount of ST2 (IL1RL1) and a decreased baseline amount of one or more of PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and / or NTN1;

[0143] iii) A ratio of baseline amounts of ST2:PASP (CPB1) has increased;

[0144] iv) A ratio of baseline amounts of ST2 to one or more of FGFBP1, SCLIBE3, SFRP1 and / or NTN1 has increased.

[0145] A baseline amount of the biomarker may reflect a measurement of the amount of biomarker in the subject before initiating an anti-cancer therapy or at the onset of an anticancer therapy. In an embodiment, the baseline for a given biomarker in the subject may refer to a measurement of the same biomarker in a population that received the same or a similar anti-cancer agent as the subject and that did not develop ILD, did not develop severe ILD, and / or that developed severe ILD, to determine whether it is relatively elevated in the subject. For example, an elevated baseline would be reflected if the subject has more of the baseline biomarker relative to any of the aforementioned populations (e.g., the population that did not develop ILD, did not develop severe ILD, or the population that developed severe ILD). A decreased baseline would be reflected if the subject has less of the baseline biomarker relative to a population that did not develop ILD or the population that did not develop severe ILD. In another embodiment, the measurement of the biomarker may occur from samples collected from the subject at two or more timepoints during treatment with an anti-cancer agent. The baseline may compare the amount of the biomarker at a later time point to a measurement of the biomarker at an earlier time point to determine if the biomarker has become elevated.

[0146] Preferably, a propensity for developing severe ILD in a subject who may be prior to receiving anti-cancer treatment, receiving an anti-cancer therapy, or beginning a cancer treatment, is determined by whether an amount of a baseline biomarker comprising ST2 (IL1RL1) has increased and, optionally whether an amount of the baseline biomarker selected from PASP (CPB1), NTN1, FGFBP1, SFRP1, and / or SCUBE3 has decreased. Preferably, the baseline biomarker is ST2 (IL1RL1) and PASP (CPB1). If the subject is above a threshold amount for the ST2 (IL1RL1) baseline biomarker, if the ratio of ST2 (IL1RL1) to PASP (CPB1) is higher in subject compared to a previous amount of ST2 in the same subject, if the ratio of ST2:PASP (CPB1) is higher than a previously determined ratio of ST2:PASP (CPB1) in the subject, and / or if the ratio of ST2 (IL1RL1) to NTN1,

[0147] 15439666-1FGFBP1, SFRP1, and / or SCLIBE3 is higher than a previously determined ratio of the same in the subject, then the subject may be identified as having a higher risk of developing severe ILD. Those individuals identified as having a higher risk of developing ILD may be further monitored for ILD using longitudinal biomarkers, receiving a prophylactic therapy for ILD such as steroids, and / or delaying, reducing, or interrupting treatment with the anti-cancer agent.

[0148] Upon determining that the subject has a higher propensity for severe ILD, then the subject’s cancer therapy and / or ILD propensity may be addressed in the following ways. The amount of anti-cancer agent being administered to the patient for treatment of the cancer may be reduced; the treatment of the anti-cancer agent being administered to the subject for treatment of the cancer may be discontinued; the severe ILD may be prophylactically treated in the subject with steroids or other prophylactic therapeutic intervention; and / or the subject may be monitored for progression or development of severe ILD.

[0149] After identifying a subject as having an increased propensity for developing severe ILD by determining an amount of baseline biomarker or ratio thereof (e.g., ST2:PASP (CPB1) or ST2 to one or more of FGFBP1, SCUBE3, SFRP1 and / or NTN1), a longitudinal analysis may be performed to monitor the subject’s risk of developing ILD. This is an example of enhanced monitoring that presently is not performed with subjects who develop ILD. It permits early intervention of I LD, to prevent or mitigate severe I LD or the risk of developing severe ILD.

[0150] Any part of the biomarker may be detected in order to determine the amount of the biomarker.

[0151] Longitudinal monitoring

[0152] All of the above-discussed biomarkers have now been found to be associated with ILD, particularly DI-ILD, and even more particularly severe DI-ILD, and it has also been found that some of the biomarkers are longitudinal biomarkers, with the biomarkers being associated with different time points before onset of ILD, and particularly with different time points before the worst diagnosis of ILD of a subject. The present disclosure, therefore, permits monitoring of a subject’s propensity for ILD by monitoring one or more biomarker in a given subject over time (e.g., by comparing amounts or ratio relative to baseline amount). For example, the amount of the biomarker in the subject may be determined at a first time point, and the amount of a biomarker in the subject may be

[0153] 15439666-1determined at a second time point, wherein the second time point is later than the first time point. The amount of a biomarker determined at the first time point is indicative of whether or not the subject has a propensity for ILD, and the amount of the same or a different biomarker can be determined at the second time point so as to monitor the subject’s propensity for ILD over time (i.e. to determine whether the subject’s propensity for ILD has increased, decreased or stayed the same over time).

[0154] In some embodiments, the biomarker at the first time point is the same as the biomarker at the second time point, and a change in the amount of the biomarker at the second time point compared to the amount of the biomarker at the first time point may indicate a change in the subject’s propensity for ILD, particularly severe ILD, depending on the direction of the change of the amount. Conversely, if there is no change in the amount of the biomarker at the second time point compared to the first time point, then it may be determined that there has been no change in the subject’s propensity for ILD.

[0155] In some embodiments, the amount of a biomarker (i.e. a first biomarker) is determined at a first time point, and the amount of a different biomarker (i.e. a second biomarker) is determined at a second time point, wherein the amounts of both the biomarker at the first time point and the biomarker at the second time point are indicative of whether or not a subject has a propensity for ILD and / or a change in the subject’s propensity for ILD. For example, if the amount of the biomarker at the first time point (i.e. the first biomarker) is indicative of a propensity for I LD and the amount of the biomarker at the second time point (i.e. second biomarker) is indicative of ILD, then it may be determined that the subject’s propensity for ILD has not changed or has increased. Conversely, if the amount of the biomarker at the first time point (i.e. the first biomarker) is indicative of a propensity for ILD and the amount of the biomarker at the second time point (i.e. second biomarker) is not indicative of ILD, then it may be determined that the subject’s propensity for ILD has not changed or has decreased. This is particularly the case when the biomarker of the second time point (i.e. the second biomarker) is associated with a shorter time to onset of ILD than the biomarker of the first time point (i.e. the first biomarker).

[0156] In order to determine whether the amount of each biomarker is indicative of ILD, the determined amount of each biomarker can be compared to an amount of the biomarker previously determined in the subject, to an amount of the biomarker in an individual not having ILD or to an index of the biomarker calculated from a plurality of individuals not having ILD.

[0157] 15439666-1In the methods of longitudinally monitoring the subject’s propensity for ILD, the amount of the first biomarker can also be determined at the second time point and can be compared to the amount determined at the first time point, wherein any change in the amount of the first biomarker between the first and second time points contributes to determining any change in the subject’s propensity for ILD. Alternatively or in addition, the amount of the second biomarker can be determined at the first time point, and the amount of the second biomarker at the second time point can be compared to the amount of the second biomarker at the first time point, wherein any change in the amount of the second biomarker between the first and second time points contributes to determining any change in the subject’s propensity for ILD.

[0158] When it is determined that a subject’s propensity for ILD has increased, a treatment for ILD can be administered to the subject, and / or an ongoing treatment can be reduced (i.e. the dosage is reduced) or discontinued. When it is determined that a subject’s propensity for ILD has not increased, or has decreased, the subject may be administered with a treatment for which the subject is in need, or the dosage of an ongoing treatment may be maintained or increased. It is therefore possible to monitor and manage the subject’s propensity for ILD, and to take prophylactic action against ILD, according to the change in the subject’s propensity for ILD over time.

[0159] In all of the methods of longitudinally monitoring the subject’s propensity for ILD, the biomarker determined at the first time point (i.e. the first biomarker) and the biomarker determined the second time point (i.e. the second biomarker) are each independently one or more of AFM, CXCL9, CXCL10, CXCL11, CXCL13, SLAMF7, SAA1, SAA2, CRP, IL-5RA, GBP1, SP-D (SFTPD), IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), PD-L1 (CD274), SFN, IL-9, HSP-90 (HSP90AB1), IL-18BP, PAPPA, FLRT3, CD87 (PLAUR), TNFAIP6, RELN, LDH-H (LDHB), ICAM1, RBP4, IL-6, AKR1A1, ITIH3, IDO1, RPS20, ILT-6 (LILRA3), CLIC3, and STAT1. Preferably, the longitudinal biomarker is one or more of SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, IL-9, IL-6, HSP-90 (HSP90AB1), TNFAIP6, RELN, AFM and RBP4.

[0160] In particular, it has been found that TNFAIP6 is upregulated 75-99 days before the worst diagnosis of ILD (Fig. 5A), that CXCL11 is upregulated 50-74 days before worst diagnosis of ILD (Fig. 5A), that each of CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), and IL-18BP is upregulated 25-49 days before

[0161] 15439666-1worst ILD diagnosis (Fig. 5A) and that each of SP-D, PAPP, CXCL9, CXCL10, PAI-2 (SERPINB2), AKR1A1, IDO1, ITIH3 and SAA2 is upregulated 0-24 days before worst ILD diagnosis (Fig. 5A). Thus, by longitudinally monitoring the amount of one or more of these proteins, it is possible to monitor a subject’s propensity for ILD, particularly severe ILD, over time. For example, if the subject shows an increased amount of one of these biomarkers at a first time point, and the same biomarker is further increased in the subject at a second, later, time point, or there is an increased amount of a different biomarker at the second time point (wherein the different biomarker may be associated to a shorter time to onset of ILD than the biomarker of the first time point), then it can be determined that the subject’s propensity for ILD has increased. As discussed above, changes in the amounts of these biomarkers can also show that the subject’s propensity for ILD has not changed or has decreased.

[0162] The following changes between the first and second time points indicate that the subject’s propensity for ILD or onset of ILD, particularly severe DI-ILD, has increased:

[0163] i) the amount of one or more longitudinal biomarkers selected from AFM, CXCL9, CXCL10, CXCL11, CXCL13, SLAMF7, SAA1, SAA2, CRP, IL-5RA, GBP1, SP-D (SFTPD), IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), PD-L1 (CD274), SFN, IL-9, HSP-90 (HSP90AB1), IL-18BP, PAPPA, FLRT3, CD87 (PLAUR), TNFAIP6, RELN, AFM, LDH-H (LDHB), ICAM1, RBP4, IL-6, AKR1A1, ITIH3, IDO1, RPS20, ILT-6 (LILRA3), CLIC3, and STAT1 is increased at the second time point compared to the first time point.

[0164] ii) the amount of one or more longitudinal biomarkers is selected from SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, RBP4, IL-9, IL-6, HSP-90 (HSP90AB1), TNFAIP6, RELN, AFM and RBP4 is increased at the second time point compared to the first time point.

[0165] iii) the amount of TNFAIP6 was increased at the first time point and the amount of one or more biomarker selected from the group consisting of CXCL11 , PAI-2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point;

[0166] iv) the amount of CXCL11 was increased at the first time point and the amount of one or more biological marker selected from the group consisting of PAI-2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point;

[0167] 15439666-1v) the amount of CXC11 is increased at the second time point compared to the first time point; and / or

[0168] vi) the amount of one or more biomarker selected from the group consisting of TNFAIP6, CXCL11, CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), and IL-18BP was increased at the first time point and the amount of one or more biological marker selected from the group consisting of SP-D, PAPP, CXCL9, CXCL10, PAI-2 (SERPINB2), AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point. In i)-vi) above, the increase at the second time point may be with respect to an amount determined at the first time point (i.e. the method comprises determining an amount of the biomarker determined at the second time point also at the first time point) or another previously determined amount of the biomarker in the subject, with respect to an amount of the biomarker in an individual not having ILD, or with respect to an index of the biomarker calculated from a plurality of individuals not having ILD.

[0169] If none of criteria i)-vi) above is met, then it may be determined that there is no increase, or that there is a decrease, in the subject’s propensity for ILD between the first and second time points.

[0170] In all of the above-described methods of longitudinally monitoring the biomarkers, it is preferred to use one or more of the following biomarkers: TNFAIP6, CXCL9, CXCL10, CXCL11, SP-D, CXCL13, SALMF7, IL-5RA and IL-18BP. TNFAIP6 and CXCL11 are even more preferred as a longitudinal biomarker.

[0171] Upon determining that a subject has an elevated risk of developing ILD based on the longitudinal biomarker analysis, the subject’s anti-cancer therapy may be adapted to reduce or eliminate the risk ILD. In an embodiment, a dosage of the anti-cancer agent being administered to the subject may be reduced to mitigate the risk of developing ILD. In an embodiment, the anti-cancer agent may be discontinued until the subject’s longitudinal biomarkers return to a non-elevated state. In an embodiment, the subject may receive a prophylactic treatment for ILD with a therapeutically effective amount of at least one steroid or other prophylactic therapeutic intervention. In an embodiment, the subject’s ILD risk may be continued to be monitored by further measurement of longitudinal biomarkers.

[0172] The method may comprise determining the amount of multiple biomarkers at each time point, for example, the amount of two, three or four, or more, biomarkers at each time

[0173] 15439666-1point. The biomarkers determined at each time point may be the same or different or may overlap. The amount of each biomarker at each time point, and / or any change in the amount of any biomarker determined between time points, indicates whether or not the subject has a propensity for ILD and / or whether the subject’s propensity for ILD has changed between time points.

[0174] Additionally or alternatively, the longitudinal method may comprise determining the amount of the biomarker(s) at more than two time points, for example, at three, four or five time points, wherein each time point is later than the preceding one. For example, the method may comprise determining the amount of a biomarker at a first time point (i.e. a first biomarker), a biomarker at a second time point (i.e. a second biomarker) and a biomarker at a third time point (i.e. a third biomarker). The biomarker at each time point may be the same biomarker at two or more of the time points, may be a different biomarker at each time point, or the biomarker may be the same for some of the time points and different for other time points. The amount of the biomarker at each time point, and / or any change in the amount of a biomarker between at least two time points, indicates whether or not the subject has a propensity for ILD and / or whether the subject’s propensity for ILD has changed between time points. It is therefore possible to monitor the subject’s propensity for ILD over time, and to adapt the treatment of the subject (e.g. increase or decrease the dosage of an ongoing treatment of the subject, discontinue an ongoing treatment of the subject, or start or re-start a treatment for which the subject is in need) in response to any changes on the subject’s propensity for ILD. For example, if it is determined that the subject’s propensity for ILD has increased at the second time point compared to the first time point, then a treatment for ILD may be administered to the subject and / or an ongoing treatment of the subject may be discontinued or the dosage thereof may be decreased. If it is then determined at a third time point that the subject’s propensity for ILD has decreased or stayed the same compared to the second time point, then the dosage of the ongoing treatment may be maintained or increased, or the discontinued treatment may be re-started, and / or the treatment for ILD may be discontinued. The amount of a biomarker may then be determined at a fourth time point to determine whether the subject’s propensity for ILD has changed compared to the third time point, and the dosage of the ongoing treatment may be maintained, increased or decreased, or ongoing treatment may be ceased, according to any change in the subject’s propensity for ILD.

[0175] Thus, the method may further comprise determining the amount of a biomarker at a third time point, which is subsequent to the second time point, and determining whether the

[0176] 15439666-1subject’s propensity for ILD has changed at the third time point compared to the second time point. It can be determined that the subject’s propensity for ILD has decreased or has not increased if:

[0177] a) the amount of one or more longitudinal biomarker selected from the group consisting of AFM, SLAMF7, SAA1, SAA2, CRP, IL-5RA, CXCL9, CXCL10, CXCL11, CXCL13, GBP1, SP-D (SFTPD), IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), PD-L1 (CD274), SFN, IL-9, HSP-90 (HSP90AB1), IL-18BP, PAPPA, FLRT3, CD87 (PLAUR), TNFAIP6, RELN, LDH-H (LDHB), ICAM1, RBP4, IL-6, AKR1A1, ITIH3, IDO1, RPS20, ILT-6 (LILRA3), CLIC3, and STATI is decreased or is not increased at the third time point compared to the amount at the second time point;

[0178] b) the amount of one or more biomarker selected from the group consisting of PAI- 2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is decreased or is not increased at the third time point compared to the second time point;

[0179] c) the amount of CXC11 is decreased or is not increased at the third time point compared to the second time point;

[0180] d) the amount of TNFAIP6 was increased at the second time point, and the amount of one or more biomarker selected from the group consisting of CXCL11, PAI-2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is decreased or is not increased at the third time point;

[0181] e) the amount of CXCL11 was increased at the second time point, and the amount of one or more biomarker selected from the group consisting of PAI-2 (SERPINB2), CAP- 3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is decreased or is not increased at the third time point; and / or

[0182] f) the amount of one or more biomarker selected from the group consisting of TNFAIP6, CXCL11, CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), and IL-18BP was increased at the second time point, and the amount of one or more biomarker selected from the group consisting of SP-D, PAPP, CXCL9, CXCL10, PAI-2 (SERPINB2), AKR1A1, IDO1, ITIH3 and SAA2 is decreased or is not increased at the third time point. Conversely, it can be determined that the subject’s propensity for ILD has increased at the third time point compared to the second time point is any one or more of criteria a)-f) above is not met.

[0183] Method of selecting a subject for treatment

[0184] 15439666-1The biomarkers also provide for a method of selecting a subject for treatment for which they are in need, wherein the treatment is associated with onset of ILD, particularly severe ILD, by allowing the subject’s propensity for ILD to be assessed. This method therefore takes place before onset of ILD.

[0185] The present application therefore provides a method of screening a subject for a treatment for which the subject is in need, wherein the treatment is associated with onset of ILD, such as a treatment comprising DXd, wherein the method comprises determining the amount of a biomarker in the subject and comparing the amount of the biomarker to an index of the biomarker. If the determined amount of the biomarker relative to the index is indicative of a propensity for ILD, then the subject may be monitored closely and may be given a prophylactic treatment that could mitigate the ILD risk. If the determined amount of the biomarker relative to the index is not indicative of a propensity for ILD, then the subject is selected for the treatment, and the method may further comprise administering the treatment to the subject.

[0186] The method may comprise determining the amount of more than one biomarker in the subject, and comparing the amount of each biomarker to a relevant index. The subject may be selected for the treatment if the amount of at least one of the biomarkers relative to the relevant index is not indicative of a propensity for I LD, or the subject may be selected for the treatment if the amount of every determined biomarker relative to the relevant indexes is not indicative of a propensity for ILD. The subject’s propensity for ILD risk as assessed by the baseline and longitudinal biomarkers may be useful to inform the potential risk and benefit of an anti-cancer therapy to the subject.

[0187] The method of screening a subject may comprise determining the amount of a biomarker in the subject and comparing the amount to an index of the biomarker, and selecting the subject for administration of the treatment if:

[0188] i) the biomarker is ST2 (IL1 RL1), and the amount of the biomarker is the same as or below the index of the biomarker;

[0189] ii) the biomarker is selected from a group consisting of PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1 , and the amount of the biomarker is the same as or above the index of the biomarker, and / or

[0190] iii) the biomarker is the ratio of ST2:PASP (CPB1), and the ratio is the same as or below the index of the biomarker. The index may, for example, be a reference amount of the biomarker derived from, or calculated on the basis of, an average amount of the biomarker calculated from a plurality of individuals not having ILD. When the biomarker in

[0191] 15439666-1the subject is more than one biomarker, then the index may be a score based on a composite value of the amount of each biomarker in a plurality of individuals not having ILD, and the score may be weighted differently for different biomarkers. Preferably, the subject is selected for the treatment when all of i)-iii) above are met. When the subject is selected for the treatment, then the method may comprise administering the treatment to the subject.

[0192] Conversely, the subject may be not selected for the treatment if one or more of i)-iii) of the preceding paragraph is not met, and the treatment is not administered to the subject. More specifically, the method of screening a subject for a treatment for which they are in need, and which is associated with onset of ILD, may comprise determining the amount of a biomarker in the subject and comparing the amount to an index of the biomarker, wherein the subject is not selected for administration of the treatment if:

[0193] i) the biomarker is selected from a group consisting of ST2, SLAMF7, SAA1 , SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, GBP1, CXCL13, SP-D, WARS1, SERPINB2, SERPINB9, SFN, LDHB, ICAM1, PD-L1, IL-18BP, RBP4, IL-9, IL-6, HSP90AB1, PAPPA, FLRT3, PLAUR, AKR1A1, ITIH3, IDO1, TNFAIP6, RPS20, LILRA3, CLIC3 and STAT1, and the amount of the biomarker is above the index;

[0194] ii) the biomarker is selected from a group consisting of CPB1 , FGFBP1 , SCUBE3, SFRP1, NTN1, RELN, AFM and RBP4, and the amount of the biomarker is below the index, and / or

[0195] iii) the biomarker is the ratio of ST2:PASP (CPB1), and the ratio is below the index. When the subject is not selected for the treatment, the treatment is not administered to the subject,

[0196] In some embodiments, the subject is selected for the treatment when:

[0197] i) the biomarker is selected from a group consisting of ST2, SLAMF7, SAA1 , SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, RBP4, IL-9, IL-6, HSP-90 (HSP90AB1), and TNFAIP6, and the amount of the biomarker is the same as or below the index;

[0198] ii) the biomarker is selected from a group consisting of PASP (CPB1), FGFBP1, SCUBE3 and SFRP1 , and the amount of the biomarker is the same as or above the index, and / or

[0199] iii) the biomarker is the ratio of ST2:PASP (CPB1), and the ratio is the same as or below the index. Preferably, the subject is selected for the treatment when all of i)-iii) above are met. When the subject is selected for the treatment, then the method may

[0200] 15439666-1comprise administering the treatment to the subject. Conversely, the subject may be not selected for the treatment if one or more of i)-iii) of this paragraph is not met, and the treatment is not administered to the subject.

[0201] In some embodiments, the subject is selected for the treatment when the biomarker is ST2 and the amount of the biomarker is the same as or below the index, the biomarker is PASP (CPB1) and the amount of the biomarker is the same as or above the index, and / or the biomarker is the ratio of ST2:PASP (CPB1) and the ratio is the same as or below the index. Preferably, the subject is selected for the treatment when all three of these criteria are met. Conversely, the subject may be not selected for the treatment if one or more of these three criteria is not met, i.e. when the amount of ST2 is above the relevant index, the amount of PASP (CPB1) is below the relevant index, and / or the ratio of ST2:PASP (CPB1) is above the relevant index.

[0202] Treatment

[0203] The methods of the present application aid the assessment of whether to start a contemplated treatment, to continue or modify an ongoing (i.e. pre-existing) treatment, or to re-start a previously discontinued treatment, wherein the subject is in need of the treatment, particularly where the treatment is associated with onset of ILD. The method also aids the assessment of whether to start a treatment for ILD, for example as a prophylactic treatment. In particular, if the determined amount of the biomarker is indicative of a propensity for ILD, then the subject can be administered with a treatment for ILD, the dosage of an ongoing treatment of the subject can be reduced or an ongoing treatment of the subject can be discontinued. It may also be decided to not administer the contemplated treatment to the subject.

[0204] Conversely, if the determined amount of the biomarker is not indicative of a propensity for ILD, or is indicative of a low propensity for ILD, then the subject may be administered a treatment for which the subject is in need, or the dosage of an ongoing treatment of the subject may be increased or maintained at the same level. This is particularly useful where the treatment is associated with onset of ILD.

[0205] In some embodiments, the treatment for which the subject is in need, or the ongoing treatment, is a treatment which is associated with the onset of ILD. In some embodiments, the treatment for which the subject is in need, or the ongoing treatment, is a treatment for cancer. In some embodiments, the treatment for which the subject is in need, or the ongoing treatment, comprises a HER2-directed antibody and / or deruxtecan. The HER2-

[0206] 15439666-1directed antibody may be an antibody-drug conjugate comprising the HER2-directed antibody, and the HER2-directed antibody may be trastuzumab. The treatment for which the subject is in need, or the ongoing treatment, is preferably an antibody-drug conjugate comprising DXd. The antibody-drug conjugate comprising DXd preferably also comprises a HER2-directed antibody, and it is particularly preferred that the antibody-drug conjugate is trastuzumab deruxtecan (T-DXd), and further details of possible treatments are disclosed in WO 2015 / 115091 which is incorporated herein by reference.

[0207] Trastuzumab deruxtecan (T-DXd) consists of trastuzumab and the drug-linker deruxtecan. Trastuzumab is a humanised monoclonal HER2-directed antibody. The amino acid sequence of the heavy chain of trastuzumab is set out in each of SEQ ID NOs: 1 and 11 , and the amino acid sequence of the light chain of trastuzumab is set out in SEQ ID NO: 2.

[0208] The drug-linker of the present disclosure includes exatecan (IIIPAC name: (1S,9S)-1-amino-9-ethyl-5-fluoro-1 ,2,3,9, 12, 15-hexahydro-9-hydroxy-4-methyl-1 OH, 13H-benzo[de]pyrano[3',4':6,7]indolizino[1,2-b]quinolin-10,13-dione, (also expressed as chemical name: (1S,9S)-1-amino-9-ethyl-5-fluoro-2,3-dihydro-9-hydroxy-4-methyl-1H,12H-benzo[de]pyrano[3',4':6,7]indolizino[1,2-b]quinolin-10,13(9H,15H)-dione)), which is a topoisomerase I inhibitor, as a component. Exatecan is a camptothecin derivative having an antitumor effect, represented by the following formula:

[0209]

[0210] The antibody-drug conjugate used in the present disclosure can be also represented by the following formula:

[0211] 15439666-1Antibody

[0212]

[0213] Here, the drug-linker is conjugated to an antibody (‘Antibody-’), in particular an anti-HER2 antibody, via a thioether bond. The meaning of n is the same as that of what is called the average number of conjugated drug molecules (DAR; Drug-to-Antibody Ratio), and indicates the average number of units of the drug-linker conjugated per antibody molecule.

[0214] After migrating into cancer cells, the above antibody-drug conjugate is cleaved at the linker portion to release a compound (which is DXd) represented by the following formula:

[0215]

[0216] When the treatment is T-DXd, T-DXd can be administered to the subject by any suitable means. For example, T-DXd may be administered to the subject intravenously ata dosage of 5.4 mg / kg or 6.4 mg / kg once every 3 weeks.

[0217] The treatment for ILD may be any known treatment for ILD. For example, the treatment for ILD may be one or more of a steroid (e.g. a corticosteroid), oxygen therapy and pulmonary rehabilitation.

[0218] Subject

[0219] Preferably, the subject has a disease or condition which is not ILD, wherein treatment of the disease or condition may be associated with onset of ILD. In some embodiments, the subject has cancer. The cancer may be breast cancer, lung cancer, gastric cancer, colorectal cancer or non-small cell lung cancer. Preferably, the subject has breast cancer or lung cancer. As discussed above, the subject may be receiving a treatment which is

[0220] 15439666-1associated with onset of ILD, preferably a treatment for cancer, which may be any of the above-discussed ongoing treatments.

[0221] In an embodiment, the subject is a mammal. Preferably, the subject is a human. In these embodiments, the individual, or each of the plurality of individuals, is a human.

[0222] Biological sample

[0223] The amount of a biomarker in a subject may be determined from a biological sample obtained from the subject. When the amount of more than one biomarker is determined, or the amount of a biomarker is determined at more than one time point, the biomarker(s) may, in each instance, be in a biological sample obtained from the subject. Similarly, the amount of a biomarker in an individual not having ILD may have been determined from a biological sample obtained from the individual. Where the determined amount of a biomarker in a subject is compared to an index of the biomarker in a plurality of individuals, the index may be derived from, or calculated on the basis of, an average amount of the biomarker determined from a biological sample obtained from each individual in the plurality of individuals not having ILD.

[0224] A biological sample obtained from the subject and / or individual(s) may be a bronchoalveolar lavage (BAL) fluid, sputum, plasma, blood, or serum sample, and is preferably a serum sample.

[0225] In embodiments where more than one biological sample is obtained from the subject, each biological sample may be a bronchoalveolar lavage (BAL) fluid, sputum, plasma, blood, or serum sample. Preferably, each biological sample is a serum sample.

[0226] Examples

[0227] Example 1. Cohort Characteristics and Research Design. All dosed and consenting patients with adjudicated drug-related ILD (n = 95) across three T-DXd clinical studies (DESTI NY-Breast01 [HER2-positive breast cancer, NCT03248492], DESTI NY-BreastO4 [HER2-low breast cancer, NCT03734029], DESTINY-LungOI [HER2-mutant non-small-cell lung cancer, NCT03505710]) were used (Figure 1A).

[0228] Clinical laboratory tests were mapped to the samples used for discovery proteomics analysis with a tolerance window of + / - 3 days. Tumor burden (sum of diameters) assessment was performed at odd cycle numbers where most samples used for proteomics analysis were from even cycle numbers. Thus, the average of 2 consecutive

[0229] 15439666-1odd visit number was used to map to the even visit number in the middle. For example, the average tumor burden at visit C3D1 and C5D1 was used to assign a tumor burden at visit C4D1. A tumor burden measurement up to 35 days prior to study start was used to assign a tumor burden at study start (C1D1). For other visits, a tolerance window of + / - 5 days was used.

[0230] Patients were classified based on their worst adjudicated ILD grade. Of the 95 patients with adjudicated drug-induced ILD included in the analysis, 74 developed mild / moderate ILD (grades 1 or 2) and 21 had severe ILD (grades 3-5) according to SomaLogic® assay, 22 in Luminex® assay (Figure 1A). Among patients who didn’t develop ILD in those cohorts, 94 patients with matching demographics (e.g., country, age, sex, best overall response, cohort) and clinical parameters were selected for SomaScan® assay and Olink® assay. The cohort was composed of individuals with an age range of 33-79 years old (mean [±SD] of 60 ± 10) and a body mass index (BMI) of 24.1 ± 4.5 kg / m2. While 100% of patients were women in DB-01 and DB-04 cohorts, 46% were male in DL-01 cohort. Longitudinal serum samples (n = 1,247) were collected at baseline (C1D1, immediately pre-treatment), within cycle 1 as well as prior to dosing across cycles 2, 3, and 4 followed by every two cycles until cycle 6 or 8, depending on the study (C2D1, C3D1, C4D1, C6D1, C8D1, etc.) (Figure 1A).

[0231] Each dosing cycle is approximately 3 weeks and 96% of samples were collected within the first 200 days of the study. On average, 6.6 ± 2.2 samples were analyzed per patient. ILD onset developed at different rates upon T-DXd treatment with the worst ILD diagnosis varying greatly with a mean [±SD] of 186 days ± 157 (Figure 1C). Median worst ILD diagnosis date was similar across DB-01, DB-04 and DL-01 cohorts with 138.5, 160.0 and 120.5 days, respectively (Figure 1 C). There was no significant difference in the timing of worst ILD diagnosis across severity grades with ILD grade 1, 2, 3 and 5 diagnosed at days 164, 150, 197 and 186, respectively.

[0232] Proteomics analysis was performed on all serum sample using 7k SomaScan® assay and a subset of the samples (i.e., DL-01 cohort) was analyzed with Olink® Explore 3072 assay. The quality of the data was examined by principal component analysis (PCA) ensuring there was no plate, cohort effect and outliers. Variance decomposition analysis further confirmed data quality with leptin identified as the top protein associated with BMI in the three cohorts. Samples, assays or proteins that did not pass acceptance criteria, as defined by the manufacturers, were discarded. SOMAScan® assay technology employs SOMAmer® (slow off-rate modified aptamer) single-stranded DNA reagents to profile

[0233] 15439666-16596 unique human proteins. The assay used 12 hybridization normalization control sequences, 5 pooled human calibrator controls, and 3 pooled quality control (QC) replicates. The QC ratio is calculated as the ratio of the QC reference value in relative fluorescence units (RFU) to the median RFU value of QC replicates for each SOMAmer® on a given plate. SomaScan® assay findings were confirmed with orthogonal technologies (i.e. Olink®) showing significant correlation on Iog2-transformed data using pearson correlation using the cor.test() function in R® or Rstudio®. Select proteins of interest were confirmed and quantified using the Luminex® assay platform with the custom Luminex® Discovery Assay Human Premixed Multi-Analyte Kit (LXSAHM-08, R&D Systems, Inc., MN, USA). Luminex® assays were expanded to include all study patients (additional 422 patients who do not develop ILD). After data curation, 6,277 SomaScan® assay probes and 2,944 Olink® proteins were retained for analysis (Figure 1A). The proteomics data were used to (1) identify baseline predictive biomarkers of ILD onset, (2) investigate mechanisms of DI-ILD and (3) discover risk-monitoring ILD biomarkers.

[0234] Example 2. Baseline ST2 predicts severe ILD onset in patients treated with Enhertu®.

[0235] Differential analysis after combination of all three cohorts identified ST2 as the most significant elevated protein in patients who developed severe ILD in comparison to those who didn’t develop ILD (Figure 2A). This trend was consistent across all three cohorts separately and remained true across the course of treatment early diagnostic biomarker (Figure 2B). The box plots shown in Figure 2B correspond to no ILD, mild ILD, and severe ILD, respectively, across all studies, DB-01, DB-04, and DL-01. The number of patients is shown along the x axis. The results were reproduced in DL-01 with Olink®. In addition, optimal cut-point analysis using the Youden index approach (using the R package “cutpointr” (v1.2.0)) identified ST2 as the most predictive marker (Figure 2C). Altogether, ST2 emerged as the most promising protein marker predisposing patients for severe ILD onset upon T-DXd treatment. Similar analyses were performed using clinical lab measurements, but none reached the same level of differences as compared to ST2. To confirm ST2 findings, a quantitative Luminex® assay was employed to quantify ST2 in the same samples but also including all other consenting non-ILD patients (n = 608) (Figure 2B). First, ST2 abundance measurements were compared across SomaScan® assay and Luminex® assay technologies and high correlation (r=.87, P-value= <2.2EA-16 for pooled samples) was observed, which confirmed the initial observations. Next, optimal cut-point analysis was performed to assess the predictive performance of ST2 using all the patients in those cohort. The cut-point is 21.4 ng / ml and was able to identify severe ILD patients quite well with AUG of 0.73, sensitivity of 0.74 and specificity of 0.73. Of note, due to the

[0236] 15439666-1low prevalence of severe ILD onset, the positive predictive value (PPV) was 9.7% meaning that -1 / 10 above ST2 threshold will develop severe ILD which corresponds to a 2.8 fold enrichment in comparison if no marker is used. Negative predictive value (NPV) was 98.8% meaning that most patients below ST2 threshold will not develop severe ILD. Patients developed severe ILD at different rates, while some patients developed severe ILD quickly within 200 days, others developed more slowly often developing lower grades first. Figure 2D shows that the speed of ILD onset is correlated with the baseline concentration of ST2 for patients who develop severe, but not mild ILD.

[0237] A significant negative correlation was observed between baseline ST2 concentration and days to worst ILD diagnosis reinforcing the relationship between ST2 level and severe ILD onset and potential biological predisposition captured with this marker. Without being bound by any particular theory, elevation in ST2 concentration may originate from neutrophils. ST2 is a systemic marker of inflammation which can be triggered by various interleukin and chemokines. ST2 levels of the present study were correlated between neutrophils count and ST2 concentration (Figure 2E). Figure 2E(i) shows that patients who develop severe ILD have heightened neutrophil levels at baseline. Neutrophil counts are correlated with ST2 at baseline and correlation is greatest in patients who develop drug induced ILD (Figure 2E(ii)). Even though inferior to ST2, neutrophil count could also classify reasonably well severe ILD patients. Figure 2F shows the cut point analysis using neutrophil count at baseline. Upon treatment, neutrophils count significantly decreased across all patients and correlation with ST2 improved in comparison to baseline samples alone, reinforcing that a sizeable portion of ST2 may originate from neutrophils (Figure 2G). Figure 2G shows longitudinal comparisons of ST2 (IL1RL1), ST2 (IL1RL1): PASP (CPB1) ratio and neutrophil counts for no ILD subjects, mild ILD subjects, and severe ILD subjects. P values are provided under the cycle and day of treatment for each graph. Altogether, ST2 may be necessary, but not sufficient, for severe ILD onset. It may be that the immunological state of patients in the context of neutrophil biology can further predispose patients for developing severe ILD.

[0238] Example 3. Baseline ST2 in conjunction with dysregulated neutrophil and fibrotic pathways predispose severe ILD onset in Enhertu®-treated patients.

[0239] As indicated above, the PPV using baseline ST2 (IL1RL1) was relatively low due to the relatively low prevalence of severe ILD. To determine what differentiates ST2 abovethreshold patients that developed severe ILD from ST2 above-threshold patients that did not develop ILD, a statistical analysis was performed among high ST2 patients and identified a set of differential proteins (see, e.g., Figure 3B) downregulated in the severe

[0240] 15439666-1ILD patient population (Figure 3A). These proteins belong to neutrophil activation as well as fibrotic pathways, indicating that dysregulation in those pathways in conjunction with high ST2 may predispose patients for severe ILD. Neutrophil activation state in conjunction with neutrophil count may improve determination of patients predisposed to severe ILD. Figure 3B shows examples of the identified differential proteins (i.e. PASP (CPB1), FGFBP1, SCUBE3, SFRP1) and their expression amounts in no ILD, mild ILD and severe ILD patient populations (the box plots correspond to no ILD, mild ILD, and severe ILD, respectively, across all studies, DB-01, DB-04, and DL-01). In each case, the biomarker was significant compared to the baseline (no ILD) and often an intermediate value was obtained in patients who developed mild ILD.

[0241] Among those differential proteins in the SomaScan® assay dataset, PASP (CPB1) was identified as the most impactful for improving the prediction of severe ILD (Figure 3C). In particular, AUG improved from 0.82 with ST2 alone to 0.90) for ST2 and PASP (CPB1) combined, with improvement in PPV from 0.32 to 0.44 (Figure 3D). Altogether, PASP (CPB1) was identified as the protein best complementing the ST2 biomarker, which improved severe ILD prediction performance, and was able to increase PPV by about 27% (Figure 3D). The complementarity of PASP (CPB1) and ST2 (IL1RL1) can be seen when taking the ratio of the two markers and the significant difference in severe ILD patients across the three cohorts (Figure 3E); the box plots correspond to no ILD, mild ILD, and severe ILD, respectively, across all studies, DB-01, DB-04, and DL-01).

[0242] Example 4. Longitudinal analysis identifies proteins and biological associated with ILD.

[0243] Linear mixed models were conducted on Iog2-transformed data at 4 time points prior to worst ILD diagnosis after time alignment of ILD diagnosis across patients using the me4’ package (v1.1 -21) in R and were adjusted for personal baseline, tumor burden, age and BMI. ‘ImerTest’ package in R (v3.1-0) was used to compute P values at each time point. P values were corrected using Benjamini-Hochberg (BH) method. Longitudinal analysis was run twice for all-grades ILD and severe ILD patients. Patients that did not develop ILD were assigned an ILD onset date at day 100. Proteins with false data rate (FDR) < 0.05 were considered significant.

[0244] Variance decomposition analysis identified tumor burden (sum of diameters), age and BMI as the main confounding factors most impacting protein variance in the dataset. Using linear mixed models following time alignment of worst ILD diagnosis across patients, and adjustment for personal baseline and the top covariates (age, body mass index, and tumor

[0245] 15439666-1burden), differential molecular trajectories in patients who developed ILD compared to those who did not were investigated.

[0246] Protein-protein network analysis was performed with “STRINGdb” (v. 2.16.4) package in R and humam database v12.0. Proteins were mapped using their Uniprot ID as provided by SomaLogic® and Olink®. Proteins with FDR < 0.05 and a change at the time point closest to ILD diagnosis of + / - 25% were used to focus on the proteins which substantial changes. Pathway enrichment analysis was performed against Gene Ontology, KEGG and Wikipathway databases across the three clusters defined using the ‘fastgreedy’ algorithm. Pathways with FDR < 0.05 were considered significant.

[0247] 595 and 510 significant proteins (FDR < 0.05) were identified in all-grade and severe ILD patients, respectively (Figure 4A). The most significant proteins were associated with both all-grade and severe ILD patients, and correlated with ILD grade, thereby indicating that differential signals were typically more pronounced in severe ILD cases. 81% of the significant proteins increased in abundance with severe ILD. Protein-protein network analysis using significant proteins (defined as FDR < 0.05 and absolute fold change > 1.25) shows that the proteins were not selected at random but were enriched in functional modules with more connections between proteins than expected (P-value < 2.33E-15).

[0248] Three clusters of proteins delineated different biological mechanisms associated with ILD onset, including inflammation, cellular damage and mRNA biology (Figure 4B). The first cluster represented the inflammatory response involving cytokine signalling (FDR = 5.67E-06) and Type II interferon signalling (FDR = 5.7E-04), leukocyte migration (FDR = 1.64E-04) and neutrophil degranulation (FDR = 6.1E-03). Interestingly, this inflammatory signature enriched for lung diseases, confirming that it originates from lung stress / damage (i.e. lower respiratory tract disease [FDR = 2.79E-07]). The second cluster resulted from cellular damage, enriching for cytosol (FDR = 2.06E-06) and mitochondrion (FDR = 7.57E-05) compartments. Finally, the third clustered was enriched with pathways associated with RNA biology including mRNA processing (FDR = 5.07E-14).

[0249] Changes of the at worst ILD diagnosis relative to four prior timepoints (log base 2 scale) were plotted for each of no ILD, mild ILD, and severe ILD patient samples (Figure 4C).

[0250] Each candidate ILD biomarker was associated with inflammation (systemic) (top row, Figure 4C), inflammation (local) (second row, Figure 4C), lung damage (third row, Figure 4C), and fibrosis (fourth row, Figure 4C). Of particular interest, markers of systemic inflammation were found to be elevated close to ILD onset, including SAA1, SAA2 and

[0251] 15439666-1CRP. The analysis also identified markers of local inflammation (i.e. CXCL9 / 10 / 11 and IL-5RA) potentially related to alveolar stress / damage (i.e. SP-D and SFN). CXCL9 / 10 / 11 are IFN-g inducible chemokines and IL-5RA is associated with eosinophilic airway inflammation. Proteins involved in pro / anti-fibrosis mechanisms were also identified, including PD-L1, IL-18BP and HSP-90 (HSP90AB1).

[0252] Altogether, the longitudinal statistical analysis efficiently captured a collection of proteins associated with ILD pathophysiology and identified leading mechanisms, including alveolar damage, inflammation and fibrosis.

[0253] Example 5. Time-course analysis identified candidate diagnostic biomarkers of ILD.

[0254] Data points were categorized in the following time windows prior to worst ILD diagnosis: 0-24, 25-49, 50-74, 75-99 days. ILD diagnosis was assigned at 128 days for non-ILD patients to maximize the number of data points in those time windows. Baseline as well as data points within the first cycle were discarded from the analysis. Statistical analysis was performed on Iog2 fold change data relative to C1D1 with Krukal-Wallis test and Dunn’s posthoc test. P-values were corrected for multiple hypothesis with BH method.

[0255] ILD is a complex disease involving multiple mechanisms including lung biology and interplay with the immune system. To identify driving mechanisms of ILD, a time-course analysis was performed that divided time by windows of 25 days prior to worst ILD diagnosis. Differential analysis in those time windows revealed that upregulation of TNFAIP6 is the first detectable event 75-99 days prior to ILD diagnosis (Figure 5A). This elevation remains significant at all subsequent three time windows. TNFAIP6 is a glycoprotein exerting anti-inflammatory properties which has been observed in the context of acute lung injury, with a role in tissue protection and immune regulation. Upregulation of TNFAIP6 likely originates from cellular stress / damages inflicted by T-DXd treatment. Even though upregulation is significant, the magnitude of changes closest to ILD onset is modest (1.4 fold increase). CXCL11 was the only protein deviating from baseline starting 50-74 days prior to ILD diagnosis, with an increase of 3.2 fold closest to ILD (Figure 5B).

[0256] Other proteins were found dysregulated up to 49 days preceding diagnosis (i.e. CAP-3 (SERPINB9), IL-6, STAT1, SAA1, IL-18BP) while others changed at the last time window likely translating into later events (i.e. SP-D (also known as “SFTPD”), CXCL9, CXCL10, etc). CXCL9 / 10 / 11 are often viewed as belonging to the same pathway and exerting similar functions. While those three proteins overall correlate well across the whole dataset, the analysis indicates that CXCL11 may precede CXCL9 and 10 changes.

[0257] 15439666-1Interestingly, earlier changes in CXCL11 were also visible along treatment cycles with significant elevation starting at C4D1 vs C6D1 for CXCL9 and CXCL10 (Figure 5C). The earlier changes can indicate driving mechanisms of ILD onset and can have value as diagnostic biomarkers. Optimal cut-point analysis indicated that CXCL11 could classify severe ILD cases vs no ILD similarly well across all three time windows, with AUG = 0.73, 0.81 and 0.77 at 0-24 days, 25-49 days, 50-74 days, respectively (Figure 5D).

[0258] Finally, key findings were confirmed using an orthogonal technology. SomaScanO-Olink® assay and SomaScan®-Luminex® assay correlations were performed on Iog2-transformed data using pearson correlation using the cor.test() function in R. Olink® Explore 3072 analysis was performed on DESTINY Lung-01 (DL-01) samples, and CXCL9, CXCL10, CXCL11, CXCL13, SLAMF7, IL-5RA, IL-18BP and SP-D correlated significantly with SomaScan® assay measurements with r > 0.78 and P-val < 2.2E-16 (Figure 5E). Select proteins were quantified with Luminex® 8-plex assay in all samples and correlation was confirmed for CXCL10 and SP-D with r> 0.85 and P-val < 2.2E-16.

[0259] Example 6. Expanded analysis of candidate diagnostic biomarker of ILD, ST2 by Luminex® assay analysis.

[0260] An expanded analysis of ST2 was carried out using the Luminex® ST2 assay. Residual serum samples from DESTI NY-BreastO2 (NCT03523585) and DESTI NY-BreastO3 (NCT03529110) were evaluated using the Luminex® assay platform with the custom Luminex® Discovery Assay Human Premixed Multi-Analyte Kit (LXSAHM-08, R&D Systems, Inc., MN, USA) as previously performed and described above. Luminex® assay analyses included all consented baseline patient samples which comprised 527 non-ILD cases, 83 mild / moderate (Grade 1 / 2), 8 severe / fatal (Grade 3-5). Patients were classified based on their worst adjudicated ILD grade. Data from DESTI NY-BreastO2 and DESTI NY-Breast03 was integrated into the previous Luminex® assay analyses for a collective total of ST2 1039 non-ILD cases, 157 mild / moderate (Grade 1 / 2) cases, 15 severe (Grade 3 / 4) cases and 15 fatal (grade 5) cases. The respective median ST2 levels in serum across the subgroups were 13.97 ng / mL in non-ILD, 12.84 ng / mL in mild / moderate ILD, 19.04 ng / mL in severe ILD, and 53.94 ng / mL in fatal cases (Figure 6A). When subgroupings were assessed collectively, a statistically significant difference was observed in fatal cases compared against all other subgroups. It is also noted that a fairly large population of patients that do not develop ILD have high levels of ST2.

[0261] When the DESTI NY-BreastO2 cohort was evaluated separately, statistically significant differences of baseline ST2 were not observed between non-ILD, mild / moderate ILD vs.

[0262] 15439666-1severe ILD cases in DESTI NY-BreastO2, though a numeric trend of increased ST2 was shown in a modest numeric median elevation observed in the study for severe ILD cases (Figure 6B), consistent with earlier studies described above. No increase in baseline ST2 was observed in the analysis of patients who developed severe / fatal ILD in DESTI NY-Breast03, however, due to low numbers of samples evaluated, the study may not be powered to support conclusions (Figure 6C).

[0263] Example 7. Concordance of analysis of candidate diagnostic biomarker of ILD, ST2 among different assays.

[0264] A concordance study of ST2 assays was carried out using the Luminex® ST2 assay and the Quantikine® ST2 assay. A bridging analysis was performed between the Quantikine® ST2 assay (results are depicted in Figure 7B) and Luminex® assay platform ST2 assay (results are depicted in Figure 7A) using patient samples from DESTINY-lungOI (NCT03505710). A strong correlation of serum ST2 levels, R=0.92, P<0.001, was observed between ILD cases evaluated (Figure 7C). The median levels of ST2 were comparable between the platforms: Non-ILD-18.13 ng / mL, mild / moderate ILD-16.45 ng / mL, and severe ILD 40.48 ng / mL in the Luminex® assay; non-ILD 12.8 ng / mL, mild / moderate 13.65 ng / mL, and 37.45 ng / mL in severe ILD. The variance and ranges across the studies were similar as well. Collectively, the Quantikine® ST2 data give confidence to the SomaScan® assay, Olink® assay and Luminex® assay assessments (described above), by giving precise outputs for ST2 with comparable outputs among the platforms. ST2 analysis between Quantikine® assay and Luminex® assay using samples from DESTINY LUNG-01 yielded an R=0.92, suggesting correlation among data, giving confidence to the previous data generated as well as for use of the Quantikine® assay when detecting ST2.

[0265] Example 8. Expanded analysis of candidate diagnostic biomarker of ILD, ST2. An expanded analysis of ST2 was carried out on samples from pan-tumor studies using the Quantikine® ST2 assay. Residual serum samples from 747 patients were evaluated from pre-treatment samples in eight clinical trials: DESTI NY-PanTumor01 (NCT04639219), DESTI NY-PanTumor02 (NCT04482309), Destiny-GastricOI (NCT03329690), Destiny-Gastric02 (NCT04014075), Destiny-Lung02 (NCT04644237), Destiny-LungOI (NCT03505710), Destiny-CRC01 (NCT03384940), Destiny-CRC02 (NCT04744831). Across the studies, 633 non-ILD cases, 87 mild / moderate (Grade 1 / 2), 11 severe (Grade 3 / 4), and 16 fatal cases (Grade 5) were evaluated using the ST2-IL33R R&D Systems ELISA kit (Quantikine®) platform. Patients were classified based on their worst adjudicated ILD grade.

[0266] 15439666-1Consistent with previous Luminex® assay analyses, a statistically significant numeric elevation in the levels of ST2 in patients who developed severe or fatal ILD (median = 27.6 ng / mL ) was observed vs. non-ILD (median= 18.8 ng / mL) & mild / moderate ILD (median= 15.4 ng / mL) across the pooled analysis of ST2 using the R&D Quantikine® platform (Figure 8A). Additionally, differences between severe ILD and fatal cases were further interrogated and identified median levels were numerically higher at the median in fatal ILD cases (median= 36.85 ng / mL) vs. severe ILD cases (median= 22.5 ng / mL) although there was a high degree of overlap in the variance and this difference was not statistically significant (Figure 8B). Given the availability of samples and incidence of events, interrogation of ST2 levels at individual study levels was difficult to perform statistical analyses. However, it is noted that of the 8 trials evaluated, Destiny-CRC01, DESTI NY-PanTumor01, and DESTI NY-PanTumor02 did not demonstrate an increase in baseline ST2 levels. Destiny-CRC01 only evaluated 3 cases of severe ILD and there is a high degree of variance in the ST2 analysis from this trial (Figure 9A-9H). The pan-tumor studies may be confounded by containing pooled analysis of N=1-2, samples from various indications such as biliary tract and rare tumors making interpretation difficult.

[0267] To further interrogate ST2 association with disease pathophysiology, clinical characteristics and risk factors associated with increased risk of adjudicated drug-related ILD were correlated with ST2 levels. Of the trials evaluated, 5 trials aggregated demonstrated a moderate correlation (R=0.3 -0.7) between the neutrophil-lymphocyte ratio (NLR) and baseline ST2 levels; and overall, baseline sST2 tends to be slightly elevated in patients with SpO2 <95% compared with >95%, suggesting an association with impaired lung function (Figure 10A-10B). Weak correlations in Destiny-GastricOI (R=0.26) and Destiny-CRC02 (R=0.2) were observed between NLR and ST2 serum levels, with Destiny-CRC01 not evaluable (Data not shown). These findings are consistent with the published Tsuchihashi etal. AACR. 2025 April 29thresults from Destiny-BreastOI, Destiny-Breast04, and Destiny-LungOI demonstrating an R=0.4,0.32, and 0.41 respectively. Higher NLR is associated with increased inflammation and generally poor prognostic outcomes. The correlation between ST2 and NLR suggests these patients may have higher rates of infection and / or inflammation prior to treatment.

[0268] Across both the Luminex® assay and Quantikine® assay analyses, subgrouping of grade 5 ILD vs. grade 3 / 4 ILD demonstrated a greater numeric increase in ILD than the pooled severe ILD subgroup. The data suggests that ST2 is associated with high grade ILD onset, with the fatal ILD cases reflecting the highest levels of ST2. Although a numeric

[0269] 15439666-1elevation of ST2 was observed in the expanded analysis of T-DXd trials, a substantial overlap exists within subpopulations of non-ILD cases. Of note, ST2 levels may vary across certain population subsets and disease, so context specific associations may be important for evaluating T-DXd drug-induced ILD. When developing assays for assessment, a PPV for any defined cut-off may be suboptimal due to a large population having high ST2 that do not develop ILD until a sufficiently large population demonstrating fatal ILD can be evaluated. ST2 analysis can be combined with other measures for increased clinical utility, such as the combinations and ratios provided herein.

[0270] 15439666-1Sequences

[0271] Amino acid sequence of the heavy chain of Trastuzumab (SEQ ID NO: 1):

[0272] EVQLVESGGGLVQPGGSLRLSCAASGFNIKDTYIHWVRQAPGKGLEWVARIYPTNGY TRYADSVKGRFTISADTSKNTAYLQMNSLRAEDTAVYYCSRWGGDGFYAMDYWGQ GTLVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSG VHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTH TCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNWYVDG VEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPP VLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

[0273] Amino acid sequence of the light chain of Trastuzumab (SEQ ID NO: 2):

[0274] DIQMTQSPSSLSASVGDRVTITCRASQDVNTAVAWYQQKPGKAPKLLIYSASFLYSGV PSRFSGSRSGTDFTLTISSLQPEDFATYYCQQHYTTPPTFGQGTKVEIKRTVAAPSVFI FPPSDEQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTY SLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

[0275] Amino acid sequence of heavy chain CDRH1 (SEQ ID NO: 3)

[0276] G F N I K D T Y

[0277] Amino acid sequence of heavy chain CDRH2 (SEQ ID NO: 4)

[0278] I Y P T N G Y T

[0279] Amino acid sequence of heavy chain CDRH3 (SEQ ID NO: 5)

[0280] S R W G G D G F Y A M D Y

[0281] Amino acid sequence of light chain CDRL1 (SEQ ID NO: 6)

[0282] Q D V N T A

[0283] Amino acid sequence of light chain CDRL2 (SAS) (SEQ ID NO: 7)

[0284] S A S F L Y S

[0285] Amino acid sequence of light chain CDRL3 (SEQ ID NO: 8)

[0286] Q Q H Y T T P P T

[0287] Amino acid sequence of heavy chain variable region (SEQ ID NO: 9) E V Q L V E S G G G L V Q P G G S L R L S C A A S G F N I K D T Y I H WV R Q A P G K G L E WV A R I Y P T N G Y T R Y A D S V K G R F T I S A D T S K

[0288] 15439666-1N T A Y L Q M N S L R A E D T A V Y Y C S R W G G D G F Y A M D Y W G Q G T L V T V S S

[0289] Amino acid sequence of light chain variable region (SEQ ID NO: 10)

[0290] D I Q M T Q S P S S L S A S V G D R V T I T C R A S Q D V N T A V A WY Q Q K P G KA P K L L I Y S A S F L Y S G V P S R F S G S R S G T D F T L T I S S L Q P E D F A T Y Y C Q Q H Y T T P P T F G Q G T K V E I K

[0291] Amino acid sequence of heavy chain (SEQ ID NO: 11) E V Q L V E S G G G L V Q P G G S L R L S C A A S G F N I K D T Y I H WV R Q A P G K G L E WV A R I Y P T N G Y T R Y A D S V K G R F T I S A D T S K N T A Y L Q M N S L R A E D T A V Y Y C S R W G G D G F Y A M D Y W G Q G T L V T V S S A S T K G P S V F P L A P S S K S T S G G T A A L G C L V K D Y F P E P V T V S W N S G A L T S G V H T F P A V L Q S S G L Y S L S S V V T V P S S S L G T Q T Y I C N V N H K P S N T K V D K K V E P K S C D K T H T C P P C P A P E L L G G P S V F L F P P K P K D T L M I S R T P E V T C V V V D V S H E D P E V K F N WY V D G V E V H N A K T K P R E E Q Y N S T Y R V V S V L T V L H Q D W L N G K E Y K C K V S N K A L P A P I E K T I S K A K G Q P R E P Q V Y T L P P S R E E M T K N Q V S L T C L V K G F Y P S D I A V E W E S N G Q P E N N Y K T T P P V L D S D G S F F L Y S K L T V D K S R W Q Q G N V F S C S V M H E A L H N H Y T Q K S L S L S P G

[0292] 15439666-1

Claims

44CLAIMS:

1. A method of predicting the propensity for, and / or monitoring for, interstitial lung disease (ILD) in a subject receiving or beginning treatment for a cancer with an anticancer agent, the method comprising:determining the amount of a baseline biomarker in the subject, wherein the baseline biomarker is one or more biomarkers selected from the group consisting of ST2 (IL1RL1), PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1; preferably wherein the baseline biomarker is selected from ST2 (IL1RL1) and PASP (CPB1).

2. The method according to claim 1 , wherein:i) an increase in ST2 (IL1RL1);ii) a decrease in the amount of one or more biomarker selected from the group consisting of PASP (CPB1), FGFBP1, SCUBE3, SFRP1, and NTN1;iii) an increase in the ratio of ST2 (IL1RL1):PASP (CPB1); and / oriv) an increase in the ratio of ST2 (IL1RL1) to any one of FGFBP1, SCUBE3, SFRP1, and NTN1;is indicative of an elevated propensity for ILD in the subject.

3. The method according to claim 2, wherein an increase or decrease in the amount of the baseline biomarker or an increase in the ratio is with respect to an amount of the baseline biomarker or the ratio previously determined in the subject, with respect to an amount of the baseline biomarker or the ratio in an individual not having ILD, or with respect to an index of the baseline biomarker or the ratio calculated from a plurality of individuals not having ILD.

4. The method according to any one of claims 1 to 3, further comprising:based on the elevated propensity for ILD, adjusting a treatment regimen for the subject by one or more of the following:i) administering a reduced amount of the anti-cancer agent, the reduced amount being less than a therapeutically effective amount of the anti-cancer agent that would have been prescribed to the subject absent the elevated propensity for ILD;ii) discontinuing treatment with the anti-cancer agent;iii) delaying treatment with the anti-cancer agent until the subject’s baseline biomarker or the ratio of baseline biomarkers is no longer increased;iv) prophylactically treating the subject for the ILD with a steroid or other prophylactic therapeutic intervention for ILD; and / or15439666-145v) monitoring the subject for progression or development of ILD.

5. The method according to claim 4, wherein the monitoring further comprises predicting an elevated propensity for ILD or an onset of ILD in the subject if one or more of:i) the amount of one or more longitudinal biomarkers selected from SLAMF7, SAA1, SAA2, CRP, IL-5RA, CXCL9, CXCL10, CXCL11, CXCL13, GBP1, SP-D (SFTPD), IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), PD-L1 (CD274), SFN, IL-9, HSP-90 (HSP90AB1), IL-18BP, PAPPA, FLRT3, CD87 (PLAUR), TNFAIP6, RELN, AFM, LDH-H (LDHB), ICAM1, RBP4, IL-6, AKR1A1, ITIH3, IDO1, RPS20, ILT-6 (LILRA3), CLIC3, and STAT1 is increased at a second time point compared to a first time point;ii) the amount of one or more longitudinal biomarkers is selected from SLAMF7, SAA1, SAA2, CRP, CXCL10, IL-5RA, CXCL9, CXCL11, CXCL13, SP-D, IFP53 (WARS1), PAI-2 (SERPINB2), CAP-3 (SERPINB9), SFN, LDH-H (LDHB), ICAM1, PD-L1, IL-18BP, IL-9, IL-6, HSP-90 (HSP90AB1), TNFAIP6, RELN, AFM and RBP4 is increased at the second time point compared to the first time point;iii) the amount of TNFAIP6 is increased at the first time point and the amount of one or more biomarker selected from the group consisting of CXCL11, PAI-2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point;iv) the amount of CXCL11 is increased at the first time point and the amount of one or more biological marker selected from the group consisting of PAI-2 (SERPINB2), CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1), IL-18BP, SFTPD, PAPP, CXCL9, CXCL10, AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point;v) the amount of CXC11 is increased at the second time point compared to the first time point; and / orvi) the amount of one or more longitudinal biomarkers selected from the group consisting of TNFAIP6, CXCL11, CAP-3 (SERPINB9), RPS20, ILT-6 (LILRA3), CLIC3, IL-6, STAT1, SAA1, GBP1, IFP53 (WARS1) and IL-18BP is increased at the first time point and the amount of one or more longitudinal biomarkers selected from the group consisting of SP-D, PAPP, CXCL9, CXCL10, PAI-2 (SERPINB2), AKR1A1, IDO1, ITIH3 and SAA2 is increased at the second time point,wherein, in i)-vi), the increase at the second time point may be with respect to an amount determined at the first time point or another previously determined amount of the15439666-146biomarker in the subject, with respect to an amount of the biomarker in an individual not having ILD, or with respect to an index of the biomarker calculated from a plurality of individuals not having ILD.

6. The method according to claim 5, wherein if the subject is predicted to have an elevated propensity for ILD, the method further comprises adjusting a treatment regimen for the subject by one or more of the following:i) administering a reduced amount of the anti-cancer agent, the reduced amount being less than a therapeutically effective amount of the anti-cancer agent that would have been prescribed to the subject absent the elevated propensity for ILD;ii) discontinuing treatment with the anti-cancer agent;iii) delaying treatment with the anti-cancer agent until the subject’s one or more longitudinal biomarkers are no longer increased;iv) prophylactically treating the subject for the ILD with a steroid or other prophylactic therapeutic intervention for ILD; and / orv) monitoring the subject for progression or development of ILD.

7. The method according to any one of claims 1 to 6, wherein the baseline biomarker is ST2 (IL1RL1) or is the ratio of ST2 (IL1RL1):PASP (CPB1).

8. The method according to any one of claims 1 to 7, wherein the longitudinal biomarker is SP-D, CXCL10, and / or CXCL11.

9. The method according to any one of claims 1 to 8, wherein:i) if the subject’s propensity for ILD is determined to be increased, then the subject is administered with a treatment for ILD, and / or the dosage of an ongoing treatment of the subject is reduced or an ongoing treatment of the subject is discontinued, orii) if the subject’s propensity for ILD is determined to be not increased, then the subject is administered with a treatment for which the subject is in need, or the dosage of an ongoing treatment of the subject is maintained or increased.

10. The method according to any one of claims 1 to 9, wherein the method is performed prior to onset of ILD in the subject.

11. The method according to any one of claims 1 to 10, wherein the ILD is severe ILD.15439666-112. The method according to any one of claims 1 to 11, wherein the ILD is drug induced-ILD.

13. The method according to any one of claims 1 to 12, wherein the amount of the baseline biomarker is in a biological sample obtained from the subject; and / or wherein each of the biomarker obtained at the first time point and the biomarker obtained at the second time point is in a respective biological sample obtained from the subject.

14. The method according to claim 13, wherein the biological sample is a plasma sample, a bronchoalveolar lavage (BAL) fluid sample, a sputum sample, a blood sample, or a serum sample.

15. The method according to claim 14, wherein the biological sample is a serum sample.

16. The method according to any one of claims 1 to 15, wherein the cancer is at least one selected from the group consisting of breast cancer, lung cancer, colorectal cancer, gastric cancer, esophageal cancer, head-and-neck cancer, esophagogastric junction adenocarcinoma, biliary tract cancer, Paget's disease, pancreatic cancer, ovarian cancer, uterine carcinosarcoma, urothelial cancer, prostate cancer, bladder cancer, gastrointestinal stromal tumor, digestive tract stromal tumor, uterine cervix cancer, squamous cell carcinoma, peritoneal cancer, liver cancer, hepatocellular cancer, corpus uteri carcinoma, kidney cancer, vulval cancer, thyroid cancer, penis cancer, leukemia, malignant lymphoma, plasmacytoma, myeloma, glioblastoma multiforme, osteosarcoma, sarcoma, melanoma, cervical cancer, uterine cancer, testicular cancer, and renal cell carcinoma.

17. The method according to claim 16, wherein the cancer is at least one selected form the group consisting of breast cancer, lung cancer, gastric cancer, colorectal cancer or non-small cell lung cancer.

18. The method according to any one of claims 1 to 17, wherein the anti-cancer agent, an ongoing anti-cancer treatment, or a treatment for which the subject is in need is treatment with an antibody-drug conjugate in which a drug-linker represented by the following formula:15439666-1wherein A represents the connecting position to an antibody via a thioether bond.

19. The method according to claim 18, wherein the antibody is an anti-HER2 antibody.

20. The method according to claim 19, wherein the antibody-drug conjugate is trastuzumab deruxtecan (DS-8201).

21. A method of screening a subject for a treatment with an antibody-drug conjugate in which a drug-linker is represented by the following formula:wherein A represents the connecting position to an antibody via a thioether bond, the method comprising:determining the amount of a biomarker in a subject and comparing the amount to an index of the biomarker, wherein if the determined amount of the biomarker relative to the index is indicative of a propensity for ILD, then the subject is not selected for the treatment, and if the determined amount of the biomarker relative to the index is not indicative of a propensity for ILD, then the subject is selected for the treatment, and15439666-149wherein the biomarker is one or more baseline biomarkers selected from the group consisting of ST2 (IL1RL1), PASP (CPB1), FGFBP1, SCUBE3, SFRP1 and NTN1.

22. The method according to claim 21, wherein the subject is selected for the treatment when:i) the amount of the baseline biomarker ST2 (IL1RL1) is the same as or below the index of the biomarker;ii) the baseline biomarker is selected from a group consisting of PASP (CPB1), FGFBP1, SCLIBE3 and SFRP1, and the amount of the biomarker is the same as or above the index of the biomarker, and / oriii) the biomarker is the ratio of ST2 (IL1 RL1):PASP (CPB1), and the ratio is the same as or below the index of the biomarker,wherein the index of the biomarker is calculated from a plurality of individuals not having ILD.

23. The method according to claim 21 or 22, wherein if the subject is selected for the treatment, the treatment is administered to the subject.

24. The method according to any one of claims 21 to 23, wherein the antibody is an anti-HER2 antibody.

25. The method according to claim 24, wherein the antibody-drug conjugate is trastuzumab deruxtecan (T-DXd).

26. The method according to any one of claims 18 to 25, wherein the anti-HER2 antibody is an antibody comprising a heavy chain comprising CDRH1 consisting of an amino acid sequence represented by SEQ ID NO: 3, CDRH2 consisting of an amino acid sequence represented by SEQ ID NO: 4 and CDRH3 consisting of an amino acid sequence represented by SEQ ID NO: 5, and a light chain comprising CDRL1 consisting of an amino acid sequence represented by SEQ I D NO: 6, CDRL2 consisting of an amino acid sequence consisting of amino acid residues 1 to 3 of SEQ ID NO: 7 and CDRL3 consisting of an amino acid sequence represented by SEQ ID NO: 8.

27. The method of claim 26, wherein the anti-HER2 antibody is an antibody comprising a heavy chain comprising a heavy chain variable region consisting of an amino acid sequence represented by SEQ ID NO: 9 and a light chain comprising a light15439666-150chain variable region consisting of an amino acid sequence represented by SEQ ID NO: 10.

28. The method according to claim 26, wherein the anti-HER2 antibody is an antibody comprising a heavy chain consisting of an amino acid sequence represented by SEQ ID NO: 1 and a light chain consisting of an amino acid sequence represented by SEQ ID NO: 2.

29. The method according to claim 26, wherein the anti-HER2 antibody is an antibody comprising a heavy chain consisting of an amino acid sequence represented by SEQ ID NO: 11 and a light chain consisting of an amino acid sequence represented by SEQ ID NO: 2.

30. A method of predicting severe ILD in a subject receiving treatment for a cancer, wherein the treatment comprises a dosage amount of trastuzumab deruxtecan, the method comprising:determining that the subject has an increased amount of a baseline biomarker comprising ST2 (IL1RL1) or a ratio of ST2 (IL1RL1):PASP (CPB1), indicating that the subject has an elevated risk for developing severe ILD;based on the elevated risk for developing severe ILD, measuring samples collected from the subject during treatment with trastuzumab deruxtecan for one or more longitudinal biomarker for ILD selected from CXC10, CXC11, and SP-D;adjusting the subject’s treatment regimen of trastuzumab deruxtecan if a later collected sample from the subject contains an elevated amount of the one or more longitudinal biomarker compared to an earlier collected sample by one or more of the following ways:i) reducing the dosage amount and / or frequency of trastuzumab deruxtecan, the reduced amount being less than a therapeutically effective amount of the anti-cancer agent that would have been prescribed to the subject absent the elevated propensity for ILD;ii) discontinuing treatment with trastuzumab deruxtecan;iii) discontinuing or delaying treatment with trastuzumab deruxtecan until a subsequently collected sample from the subject exhibits a decrease or comparable amount of the longitudinal biomarker relative the later or the earlier sample of the subject, the subsequently collected sample being collected after the later sample; and / or15439666-151iv) prophylactically treating the subject for the ILD with a steroid or other prophylactic therapeutic intervention for ILD.

31. A method of predicting the propensity for, and / or monitoring for, interstitial lung disease (ILD) in a subject receiving or beginning treatment for a cancer with an anticancer agent, the method comprising:determining the amount of a baseline biomarker in the subject, wherein the biomarker is ST2 (IL1 RL1),wherein an increase in ST2 (IL1RL1) is indicative of an elevated propensity for ILD in the subject,wherein the increase in the amount of the baseline biomarker is with respect to an amount of the baseline biomarker previously determined in the subject, with respect to an amount of the baseline biomarker in an individual not having ILD, or with respect to an index of the baseline biomarker calculated from a plurality of individuals not having ILD,wherein i) if the subject’s propensity for ILD is determined to be increased, then the subject is administered with a treatment for ILD, and / or the dosage of an ongoing treatment of the subject is reduced or an ongoing treatment of the subject is discontinued, orii) if the subject’s propensity for ILD is determined to be not increased, then the subject is administered with a treatment for which the subject is in need, or the dosage of an ongoing treatment of the subject is maintained or increased,wherein the ILD is severe drug induced-ILD,wherein the subject has cancer,wherein the cancer is at least one selected form the group consisting of breast cancer, lung cancer, gastric cancer, colorectal cancer and non-small cell lung cancer, andwherein the ongoing treatment or the treatment for which the subject is in need is treatment with trastuzumab deruxtecan (T-DXd).15439666-1

Images