Diagnosis of haemolysis

Abstract

The invention provides methods for determining the presence or absence of haemolysis in a subject, as well as devices suitable for use in these methods.

Description

[0001] DIAGNOSIS

[0002] Field of the Invention

[0003] The invention relates to methods and kits for detecting haemolysis in a subject. The invention also allows to distinguish haemolysis, in which ruptured red cells release their contents into the bloodstream (e.g. intravascular haemolysis), from blood contamination of a sample, and further is useful in monitoring disease.

[0004] Background of the Invention

[0005] Red blood cells (RBCs) are present in every perfused tissue and account for one in five human cells in the body. Circulating RBCs are vulnerable to rupture (haemolysis, e.g. intravascular haemolysis), resulting in the release of their contents into the plasma. Higher levels of intravascular haemolysis may indicate the presence of triggers, including pharmacological, biochemical, infective, immunological or mechanical. Elevated levels of RBC destruction can lead to extravascular haemolysis, also releasing contents. Excessive extravascular haemolysis may occur in inherited anaemias that produce abnormal RBCs that are destroyed in the liver, spleen or bone marrow. Some conditions increase the propensity of rupture. The timely detection of haemolysis is critical because consequences can be severe, including anaemia and adverse reactions to released cellular contents. Due to the abundance of RBCs, haemolysis can be substantial. Haemolysis is, therefore, universally recognised as an adverse event that must be monitored to inform clinical decisions. However, at present, diagnosing haemolysis typically requires blood sampling, which involves significant staff, equipment, and consumables costs. Moreover, the collection and subsequent processing of blood samples can induce RBC rupture outside the body, for example as blood is drawn through a needle under suction, when performing a centrifugation step, or during storage of the sample prior to processing, resulting in a false-positive results. In addition, said tests cause discomfort to patients, which limits the sampling frequency and practically restricts testing to hospital admissions or clinic appointments.

[0006] The current tests or surrogates of haemolysis have a number of limitations such as: providing indirect readings that cause delays in care decisions (bilirubin), requiring calibration, a baseline measurement and contextualization to the health or maturity state (haptoglobin; Hp), involving complex biochemical assays (lactate dehydrogenase; LDH), detecting only a single cause (direct Coombs test), or not being specific to haemolysis or its causes therefore yielding false-positives (haemoglobin; Hb). In more detail, reticulocyte count increases after haemolysis, but this response is slow -onset and non-specific to haemolysis (Aladjidi N. et al. 2011, Haematologica, 96(5), pp. 655-663; Liesveld J.L. et al., 1987, Blood, 69(3), pp. 820-826.). Similarly, elevated serum LDH is not necessarily attributable to ruptured RBCs (Mecozzi G. et al., 2002, JThorac Cardiovcisc Surg, 123(3), pp. 550-556; Pourrat O. et al., 2015, Eur J Obstet Gynecol Reprod Biol, 789, pp. 68-72).

[0007] Hb levels are, arguably, the most direct indicator of clinical severity of haemolysis (Barcellini W. et al., 2014, Blood, 124(19), pp.2930-2936; Michel M. et al., 2009, Blood, 114(15), pp. 3167-3172.), but interpretation hinges on knowing prior levels or at least a time course. Moreover, low haemoglobin may indicate suppressed erythropoiesis (e.g. iron deficiency anaemia) rather than increased RBC rupture. Blood levels of Hp, a glycoprotein made by the liver (Bowman B.H. and Kurosky A., 1982, Adv Hum Genet, 12, pages 189-261, 453-454.), decrease upon haemolysis due to binding with free Hb, but accurate interpretation is, again, contingent on knowing baseline levels and the test’s resolving power is constrained by being an inverse indicator i.e. requiring subtraction (Shih A.W. et al., 1989, Am JHematol, 89(4), pp. 443-447.; Marchand A. et al., 1980, JAMA, 243(19), pp. 1909-1911.). Moreover, Hp levels are influenced by factors unrelated to haemolysis, such as liver maturation.

[0008] Bilirubin measurements, favoured in neonatal care, describe the downstream catabolism of the haem group and therefore provide a delayed readout of a haemolytic event, although more strongly associated with extravascular haemolysis (Berlin N.I. and Berk P.D., 1981, Blood, 57(6), pp. 983-999). Conveniently, bilirubin levels can be inferred using skin bilirubinometry, although this is inaccurate with darker skin tones and false-positives can arise when hepatic clearance of conjugated bilirubin is inadequate or in conditions such as Crigler-Najjar Syndrome. False-positives are notably problematic in direct antiglobulin test (DAT) in neonates (Berentsen S., 2018, Br J Haematol, 181(3), pp.320-330; Gehrs B.C. and Friedberg R.C., 2002, Am JHematol, 69(4), pp. 258-271.; Parker V. and Tormey C.A., 2017, Arch Pathol Lab Med, 141(2), pp. 305-310), where 30-50% of positive cases may not undergo haemolysis.

[0009] It is clear that current methods can produce indirect, ambiguous, poorly calibrated, or delayed results that rely on adequate resources. Rapid and resource-efficient methods, compatible with minimal-laboratory or point-of-care testing, would enable routine monitoring and expedited decisions based on clinically actionable, real-time information.

[0010] As alternatives to blood testing, urinalysis methods have been introduced for markers such as hemosiderin (Sears D.A. et al., 1966, Blood, 28(5), pp.708-725). Whilst attractive from the viewpoint of cost-effectiveness and accessibility, low test sensitivity is a problem because haemosiderinuria develops once the binding capacity of Hp is saturated, a process that may take several days.

[0011] There is an unmet need to provide an improved method for assessing haemolysis using non- invasive tests, which have the option of being performed outside clinical settings (e.g. at home), without additional processing of collected samples. In healthcare facilities, there is an unmet clinical need for frequent testing in patients with genetic (glucose-6-phosphate dehydrogenase deficiency [G6PDD], pyruvate kinase deficiency [PKD], inherited anaemias e.g. sickle cell disease [SCD]) or environment-related (Epstein-Barr, malaria, dengue) risks. In addition, haemolytic reactions to transfusion or bone marrow transplant — albeit rare — can have costly complications, avoidable through monitoring.

[0012] The invention provides improved methods and kits that address these problems.

[0013] Summary of the preferred embodiments

[0014] The invention provides methods for determining the presence or absence of haemolysis inside blood vessels (such as intravascular haemolysis) in a mammalian subject. It comprises a step of detecting carbonic anhydrase 1 (CAI) in a urine sample from the subject, wherein the subject is not a human neonate. These methods are particularly advantageous as they allow real-time detection of haemolysis (such as intravascular haemolysis) with high accuracy and thus provide an improved diagnosis. Furthermore, as they use a non-invasive sample, they are easy to administer even outside a clinical setting or in the absence of specialised equipment.

[0015] In some embodiments, the subject is a human, a dog, a cat or a horse. For example, dogs are often used in sport and thus are prone, for example, to exertion -induced haemolysis. Likewise, horses (in particular racehorses) are used for sport and a lack in performance due to haemolysis would be particularly detrimental.

[0016] Where the subject is a human, the human may be a child or an adult. The human may be at least 29 days old, at least 1 year old, at least 5 years old, at least 10 years old, at least 15 years old or at least 18 years old.

[0017] The invention also provides a method for determining the presence or absence of haemolysis inside blood vessels (such as intravascular haemolysis) in a human neonate who does not meet the threshold for phototherapy according to the National Institute for Health and Care Excellence (NICE) guidelines, comprising a step of detecting CAI in a urine sample from the human neonate. The invention thus allows haemolysis to be detected in neonates which would previously not have been diagnosed with haemolysis and thus not been treated accordingly. The invention thus allows for early intervention in neonates even where standard methods would not have indicated treatment.

[0018] The non-invasive sample is a urine sample. This is preferred as the inventors have found that the analysis of urine samples is particularly useful in the context of the invention because such samples are easy to obtain and so can even be used outside of clinical settings. In alternative embodiments the methods of the invention may also be practised with another sample such as blood, plasma or serum.

[0019] The methods of the invention may further comprise a step of quantifying CAI.

[0020] In some embodiments, the method further comprises a step of detecting haemoglobin (Hb). This is useful as it may highlight a false-positive result arising from contamination of samples by blood, for example, bleeding along the urogenital tract.

[0021] In some embodiments, the method comprises quantifying the marker Hb. Quantification of CAI or Hb can be done using proteomics or an immunoassay using binders such as antibodies or aptamers, synthetic antibodies and nucleic acid or small molecule binders suitable for detecting CAI or Hb. Suitable antibodies, specific binders or probes will be known to a skilled person and are commercially available. Alternatively, Hb may be quantified using a photoelectric colorimeter measuring absorbance, using a haemoglobin colour scale, or by spectrophotometry. These methods would be known to a skilled person. In the methods of determining the severity of haemolysis in a neonate, the markers are preferably quantified as this allows for a more precise assessment.

[0022] In some embodiments, the method further comprises a step of comparing the concentration of the one or more marker(s) to a reference.

[0023] In some embodiments, the marker is quantified using an ELISA assay and / or a western blot assay and / or a lateral flow immunoassay.

[0024] The invention also provides a method of distinguishing between haemolysis inside blood vessels (such as intravascular haemolysis) and blood contamination of a sample, comprising a step of detecting CAI and Hb. The presence of CAI and absence of Hb would indicate haemolysis proximal to the glomerular barrier, whereas the presence of CAI and presence of Hb would indicate blood contamination of the sample, a false-positive diagnosis of haemolysis inside blood vessels. In some embodiments, the subject is a human suffering from a disease associated with haemolysis. The disease may be selected from the group consisting of an infection, reaction, enzyme deficiency, inherited anaemia, and adverse biochemical, pharmacological or immune reactions.

[0025] Where the disease is a bacterial infection, haemolysis can be triggered by bacterial invasion into RBCs (e.g. Bartonella causing Carrion’s disease), production of toxins (e.g. Clostridium perfringens, enterohaemorrhagic strains of Escherichia coli, Shigella dysenieriae). activation of immune system (e.g. Mycoplasma pneumoniae, Rickettsia rickeiisii. Salmonella enterica) or by triggering of haemolytic uraemic syndrome (HUS) (e.g. Streptococcus pneumoniae).

[0026] Where the disease is an infection caused by RNA or DNA viruses, haemolysis can be triggered by viruses which can adhere to and infect RBCs (e.g. West Nile virus), target erythroid progenitor cells in the bone marrow (e.g. Parvovirus B19), or cause autoimmune haemolytic anaemia (e.g. Ebstein -Barr virus, human immunodeficiency virus, morbillivirus, paramyxoviruses, rubivirus, varicella-zoster virus, cytomegalovirus, hepatitis A virus, dengue virus and large group of viruses of Flaviviridae, Filoviridae, Arenaviridae or Bunyaviridae families causing haemorrhagic fevers).

[0027] Where the disease is a protozoan or parasitic infection, haemolysis can be cause by direct invasion or binding to RBC (e.g. Plasmodium causing malaria, Trichomonas vaginalis causing trichomoniasis, Leishmania causing leishmanioses, Trypanosoma causing trypanosomal disease, Babesia causing babesiosis), production and release of toxins or proteases (e.g. Entamoeba histolytica, Trichomonas vaginalis), activation of immune system (e.g. trypanosomal diseases) by inducing HUS (e.g. Entamoeba histolytica).

[0028] Where the disease is an enzyme deficiency, it may be selected from the group consisting of glucose 6-phosphate dehydrogenase deficiency (G6PDD), pyruvate kinase deficiency (PKD), and glutathione synthetase deficiency (GSD).

[0029] Where the disease is an inherited anaemia, it may be selected from the group consisting of thalassemia, sickle cell disease, hereditary spherocytosis, and hereditary elliptocytosis.

[0030] Where the disease is an adverse reaction, it may be selected from the group consisting of HEEEP syndrome, thrombotic thrombocytopenic purpura (TTP) and haemolytic uremic syndrome (HUS), and rhesus disease. In some embodiments, the subject suffers from exertion -induced haemolysis which may have been caused by mechanical damage arising from strenuous exercise.

[0031] In some embodiments, the subject suffers from haemolysis triggered by an adverse reaction to food and beverage intake, including, blueberries, alcohol, legumes (such as fava beans, peanuts, peas and soybeans), products with artificial dyes, containing menthol.

[0032] In some embodiments, the subject suffers from haemolysis triggered by toxins or poisons such as Acalypha indica poisoning.

[0033] In some embodiments, the subject suffers from haemolysis triggered by an adverse reaction to a pharmaceutical agent, such as some antibiotic (e.g. penicillins, cephalosporins, sulphonamides, rifampin), anti-inflammatory (e.g. ibuprofen, paracetamol), anti-malarial (e.g. primaquine, quinine, quinidine), or chemotherapeutic (e.g. cisplatin, mitomycin) drugs.

[0034] The invention also provides a method of determining the presence or absence of an infective, immunological, pharmacological, biochemical, food or beverage trigger of haemolysis, comprising a step of detecting CAI in a urine sample from the subject.

[0035] The invention also provides a method for determining severity of haemolysis in a mammalian subject, comprising a step of detecting CAI in a urine sample from the subject.

[0036] The invention also provides a method for determining the efficacy, safety and / or dose-escalation of a drug treatment in a mammalian subject, comprising a step of detecting CAI in a urine sample from the subject.

[0037] In some embodiments, the patient suffers from a disease selected from the group consisting of Bartonella causing Carrion’s disease, Clostridium perfringens, enterohaemorrhagic strains of Escherichia coli, Shigella dysenteriae, Mycoplasma pneumoniae, Rickettsia rickettsii, Salmonella enterica, Streptococcus pneumoniae, West Nile virus, Parvovirus B 19, Ebstein -Banvirus, human immunodeficiency virus, morbillivirus, paramyxoviruses, rubivirus, varicellazoster virus, cytomegalovirus, hepatitis A virus, dengue virus, Flaviviridae, Filoviridae, Arenaviridae or Bunyaviridae, Plasmodium causing malaria, Trichomonas vaginalis causing trichomoniasis, Leishmania causing leishmanioses, Trypanosoma causing trypanosomal disease, Babesia causing babesiosis, Entamoeba histolytica, sickle cell disease, thalassemia, hereditary spherocytosis, hereditary elliptocytosis, glucose 6-phosphate dehydrogenase deficiency (G6PDD), pyruvate kinase deficiency (PKD), glutathione synthetase deficiency (GSD), rhesus disease, HELLP syndrome, thrombotic thrombocytopenic purpura (TTP) and haemolytic uremic syndrome (HUS). Preferably, the patient suffers from malaria.

[0038] In some embodiments, the method further comprises quantifying CAI.

[0039] In some embodiments, the method further comprises a step of comparing the concentration of CAI to a reference value.

[0040] The invention provides a method of distinguishing between haemolysis inside blood vessels (such as intravascular haemolysis) and blood contamination of a sample, comprising a step of detecting CAI and Hb in the sample.

[0041] In some embodiments, the presence of CAI and absence of Hb indicates haemolysis proximal to the glomerular barrier or the presence of CAI and presence of Hb indicates blood contamination.

[0042] In some embodiments, CAI is quantified using an ELISA assay and / or a western blot assay and / or a lateral flow immunoassay.

[0043] In some embodiments, a concentration above the reference value indicates the presence of haemolysis, or a concentration equal to or below the reference value indicates the absence of haemolysis. “Equal to" in this context may mean that the value is within a range of ± 5%.

[0044] In some embodiments, the method is performed ex vivo or in vitro.

[0045] The invention also provides a kit comprising two or more antibodies wherein at least one antibody can detect CAI and at least one antibody can detect Hb. In some embodiments, the two or more antibodies are monoclonal, polyclonal or recombinant antibodies. Preferably, the kit comprises a lateral flow immunoassay. This is useful as such tests can easily be used at home and thus allow the diagnosis of haemolysis outside clinical facilities, such as at home.

[0046] The invention also provides a lateral flow assay device comprising (i) a sample receiving region; and (ii) a capture membrane positioned downstream of the sample receiving region and comprising a test region including an antibody suitable for detecting CAI . The capture membrane may further comprise an antibody suitable for detecting Hb.

[0047] The invention also provides the use of the kit or lateral flow assay device as described herein in a method of determining the presence or absence of haemolysis in a subject. In some embodiments, the haemolysis is intravascular haemolysis or extravascular haemolysis or bleeding along the urogenital tract.

[0048] Brief Description of the Drawings

[0049] Figure 1: Study design. (A.) Four arms of the study, showing breakdown by age (years) or gestational age (weeks) among male and female participants (no significant differences). (B.) Breakdown of participants in Bangladesh: healthy controls, uncomplicated malaria (Uncomp), and complicated malaria (Comp). (C.) Breakdown of participants by urinalysis blood score (0- 4) along y-axis and haemolysis score (0-12), color-coded from black (negative) to grey, for patients with and without Foley catheter. (D.) Breakdown of participants in Peru by health center: Santo Tomas [ST], San Jose de Lupuna [SL], Santa Clara [SC] and Varillal [V]. Black bars indicate participants who tested positive for malaria infection. (E.) Recruitment and procedures on neonates. For each participant, column one: urine samples (grey square); column two: serum and / or transcutaneous bilirubin (black square); column three: blood test (Hb, full blood count, CRP, DAT; black circle).

[0050] Figure 2: Urinary CAI immunoreactivity. (A) Illustration of the mechanism of CAI appearance in urine. (B). Urine CAI or Hb-alpha EEISA quantification, normalized to positive control (10,000x diluted blood lysate). Thresholds defined using the Oxford reference cohort after removing outliers identified by Grubb’s method. OXREF: Oxford reference samples; OXHOSP: samples from Oxford hospital admissions; BD: Bangladeshi samples; PE: Peruvian samples; LO: London samples. Relationship between ELISA CAI and ELISA Hb quantification on a log-log plot. Samples were divided into four groups using thresholds defined for CAI and Hb. CAl-positive / Hb-negative samples are highly likely haemolysis inside blood vessels (bottom right square). Dually negative samples are likely non-haemolyzing participants (bottom left square). Samples where CAI signal exceeds Hb signal by at least 33% were deemed to have at least a component of haemolysis (right-hand triangle of the top right square). The remaining samples of comparable Hb and CAI signal were deemed to have been contaminated with bleeding into urine (left-hand triangle of the top right square). (C.) Breakdown of samples by number or (D.) by percentage according to classification. (E.) Western blots for human albumin (hAlb), carbonic anhydrase 1 (CAI) and haemoglobin alpha (Hb-alpha) in the Oxford, Bangladeshi, Peruvian and London cohorts. Also shown is a concentration-response for RBC lysates and recombinant human CAI (rhCAl) diluted in CAl-free adult urine. Adult samples were taken on one occasion; neonatal samples were taken longitudinally on the days indicated, over the first 10 postnatal days of life. (F.) Relationship between CAI densitometry and its visual score (0-3) across the five study groups. Color-coding relates to the visual score for each western, from not detectable (score of zero) to highest intensity (score of 3). (G.) CAI ELISA quantification, normalized to positive control (10,000x diluted blood lysate), versus western blot visual score. Summary statistics show mean±SEM.

[0051] Figure 3: Urinary CAI excretion in sickle cell disease and other inherited anaemias. Urinary CAI excretion, measured by (A-B) ELISA or (C.) western in sickle-cell (SCD) and non- haemolysing anaemias, shown in comparison to the reference group of healthy participants. (D) Blood Hb, (E) reticulocytes (rets), (F) LDH, (G) bilirubin, (H) blood Hb, (I) LDH, (J) bilirubin, (K) reticulocyte (rets) count, in sickle-cell patients (SCD; light grey) and non-hemolyzing anemic patients (Hereditary spherocytosis, Hemochromocytosis, Diamond Blackfan anemia; darker grey). (L.) Correlation (Pearson’s test) between urinary CAI ELISA signal and blood Hb, bilirubin, LDH and reticulocyte count. (M) Results of correlation analyses.

[0052] Figure 4: Urinary CAI excretion is a sensitive indicator of haemolysis. (A.) Blood Hb, plasma free Hb, LDH, total bilirubin and creatinine levels in uncomplicated and complicated cases of malaria in the Bangladeshi cohort. Light grey box (above the dashed line in the first graph; below the dashed line in the other four graphs) indicates reference range. Dashed line indicates threshold that best separates the two populations (maximal Youden index after ROC analysis). (B) Relationships between urinary CAI ELISA and patient group. Lighter color indicates higher corresponding urinary Hb ELISA signal. (C.) Box plot for CAI ELISA by patient group. (D.) Relationship between urinary CAI score (western blot) and patient group. Note, the ELISA data are coded in proportion to Hb ELISA signal. U=uncomplicated malaria, CACrL=complicated malaria with low creatinine, CACrH=complicated malaria with high creatinine. (E.) Relationship between serum LDH and urinary CAI ELISA from the uncomplicated (light grey) and complicated (dark grey) malaria cases for samples with normal kidney function (Cr<2.2 mg / dL). Lines indicate 95% confidence interval of best fit. (F.) Relationship between plasma free Hb and urinary CAI ELISA from the uncomplicated (light grey) and complicated (dark grey) malaria cases for samples with normal kidney function (Cr<2.2 mg / dL). Lines indicate 95% confidence interval of best fit. (G.) Decision tree analysis against urinary CAI western blot score identified plasma free Hb as the most important predictor. (H.) Effect size and significance of the predictors, highlighting plasma free Hb (interactions with urine protein score and normal / abnormal color) as the most important predictors of urinary CAI excretion. (I.) Regression analysis produced a highly significant model for estimating CAI ELISA signal. (J.) Urinary CAI ELISA signal in patient sub-groups broken down by plasma free Hb (threshold 0.07 g / L) and further by normal or abnormal color, as determined by decision tree analysis. Symbol color denotes Hb ELISA signal (lighter, higher Hb). (K.) Data from J grouped by patient type: healthy (black), uncomplicated malaria (dark grey), complicated malaria with normal renal function (Cr<2.2; light grey) and complicated malaria with renal failure (Cr>2.2; light grey with black outline). (L.) Relationship between plasma free Hb (threshold 0.07 g / L) and CAI western blot score (negative / positive) for patients broken down by group, healthy (black), uncomplicated malaria (dark grey), complicated malaria with normal renal function (Cr<2.2; light grey) and complicated malaria with renal failure (Cr>2.2; light grey with black outline)

[0053] Figure 5: Urinary CAI excretion as a marker of inflammatory state. (A.) Blood Hb, haptoglobin, direct bilirubin, indirect bilirubin, and CRP in malaria positive and negative diagnoses in the Peru cohort. Light grey box indicates reference range. Dashed line indicates threshold that best separates the two populations (maximal Youden index after ROC analysis). Summary statistics show mean±SEM. Significance tested by t-test. (B.) Euler diagram showing the classifications of Peruvian participants by CRP status, malaria diagnosis, malaria medication, and fever. (C.) Result of decision tree showing greatest influence of CRP on urinary CAI western blot score. (D.) Relationship between demographic or clinical information and urinary CAI score. Significance testing by Spearman’s correlation. (E.) Relationship between blood measurements urinary CAI score. Significance testing by Spearman’s correlation. (F.) Relationship between CRP and urinary CAI score. Symbols encode participant status (control: white circle, fever without malaria: light grey circle, malaria without medication: dark grey circle, malaria with medication: black star). (G.) Summary statistics for relationship between CRP and CAI western score: (G.) Thresholding CRP at 22 mg / L confirms >90% specificity and sensitivity of the test (control: white, fever without malaria: light grey, malaria without medication: dark grey, malaria with medication: black). (H.) Breakdown of participants by CRP and disease status for CAI ELISA result (control: white circle, fever without malaria: light grey circle, malaria without medication: dark grey circle, malaria with medication: black star)

[0054] Figure 6: Postnatal urinary CAI excretion categorizes newborn and associates with infection markers. (A.) Quantification of urinary CAI excretion in the first 10 days of life in 32 neonates, grouped into three patterns. Two participants were unclassified. Intensity of square symbols relates to ELISA CAI score, from zero (white) to highest (black). Intensity of circle symbol relates to western blot densitometry, from zero (white) to highest (black). Ranking within group is by mean ELISA CAI level. Also shown mean bilirubin level normalized to the NICE threshold for phototherapy; peak CRP during the 10-day period; early WBC and Hb are the first records; late WBC and Hb are mean of subsequent measurements; DAT is result of Direct Antiglobulin Test. Bottom: clinical reason for admission. CONGE: congenital condition; PREMA: premature birth; PNEUM: pneumothorax; STROK: stroke; HYPOT: hypotonia. Intensity of symbols is proportional to metric amplitude. (B.) Time courses of CAI ELISA results in the three haemolysis groups. (C.) Time courses of bilirubin. Dashed line shows threshold for indicating phototherapy according to NICE guidelines. Mean / SEM for available data. (D.) Principal component analysis of standardized data by group: physiological (white), transient (grey) and sustained (black). (E.) Gestational age by group: physiological (white), transient (grey) and sustained (black). Circle indicates full-term baby; star indicated pre-term baby. (F). Peak CRP and (G) mean white blood cell count (WBC) by group (see E).

[0055] Figure 7: Lateral flow device for urinary CAI. (A.) Exemplar LFD test results from the five participant cohorts plus calibration standards and densitometric analysis of their T and C bands. (B.) T / C ratios and calibrations for the first and (C.) second LFD design, varying in terms of antibody concentration. (D.) Assessment of sensitivity and specificity of LFD reading against laboratory CAI tests. TP=true positive, TF=true negative, FP=false positive, FN=false negative.

[0056] Detailed Description of the preferred embodiments

[0057] Haemolysis

[0058] Haemolysis is the breakdown of RBCs and the release of their contents into the surrounding fluid. Haemolysis is part of the normal functioning of the body, as old or faulty RBCs are destroyed in the spleen and elsewhere in the body. Pathologically, haemolysis can occur due to abnormally high levels of RBC rupture inside blood vessels or due to abnormally high levels of RBC destruction in the spleen, liver or bone marrow by macrophages.

[0059] Thus, haemolysis can be categorised into intravascular haemolysis and extravascular haemolysis. Intravascular haemolysis occurs within the blood vessels, releasing the contents of the RBCs, including CAI into the plasma, and reflects an adverse reaction to infective, immunological, biochemical or pharmacological trigger that may be exacerbated by a genetic predisposition. In contrast, extravascular haemolysis takes place outside of the vascular system, for example in the liver, spleen, bone marrow or lymph nodes and may also release contents, including CAI. As used herein, the term “haemolysis inside blood vessels” includes intravascular haemolysis events which occur in circulating red blood cells, and extravascular haemolytic events which result in the release of contents, including CAI, into the blood stream. The invention provides methods and kits that allow for the detection of haemolysis in a subject. These methods are based on the inventors’ discovery that urinary excretion of carbonic anhydrase 1 (CAI) is a good biomarker of haemolysis taking place proximal to the glomerular barrier. After haemoglobin (Hb) isoforms, CAI is the most abundant protein in RBC cytoplasm (~4 g / L of blood in adults ~1 g / L of blood in newborn). Thus, RBC rupture can lead to a significant rise in plasma CAI levels, normally kept at negligible levels (Wahlstrand T. et al., 1979, Scand J Clin Lab Invest, 39(6), pp.503-509.; Backman U. et al., 1990, Scand J Clin Lab Invest, 50(6), pp. 627-33.). Unlike Hb, the CAI molecule does not bind to plasma proteins (as the case for haptoglobin binding Hb), is sufficiently small to pass the glomerular filter freely and is not recycled or reclaimed by tubular mechanisms (Maren T.H., 1967, Physiol Rev, 47(4), pp. 595-781; Sly W.S. and Hu P.Y., 1995, Annu Rev Biochem, 64:375-401; Supuran C.T., 2008, Nature reviews Drug discovery, 7(2), pp. 168-181; Robinson J.R., 1950, Journal of clinical pathology, 3(2), pp.142-145.).

[0060] Conveniently, CAI presents with epitopes that are conducive for immunodetection using assays including western blotting, ELISA and lateral flow immunoassay.

[0061] The inventors have thus provided methods that can detect and quantify haemolysis in a urine sample from a subject with high accuracy. The methods comprise a step of detecting CAL These methods can easily be carried out using a lateral flow immunoassay. Such a test is easy to interpret and could be performed at home as often as deemed necessary. A positive test can instigate lifestyle changes, such as withdrawing a potential trigger (e.g. food, drug, or activity), or give a compelling reason for timely hospital admission. Regular testing could help doctors decide how best to treat a patient and flag adverse reactions early to avoid costly complications. The prospect of home-based testing could encourage early discharge after an intervention, such as surgery, blood transfusion, or a new drug regimen. Globally, the device can enable diagnosis where blood testing has been unfeasible (e.g. malaria).

[0062] The invention provides methods for determining the presence or absence of haemolysis in a urine sample from a mammalian subject. The methods of the invention are particularly useful because they allow for the accurate diagnosis of haemolysis even in patients in which haemolysis could not be detected otherwise. This is achieved by detecting CAI in a urine sample. The method may comprise a further step of detecting haemoglobin (Hb), the presence of which would indicate blood contamination of the sample (i.e. a false-positive). The methods of the invention allow for rapid and resource -efficient methods, compatible with minimal- laboratory or point-of-care testing, enables routine monitoring and expedited decisions based on clinically actionable, real-time information. Said methods are able to quickly and cheaply detect haemolysis inside blood vessels, enabling monitoring as to whether the patient is experiencing an adverse reaction to a range of haemolytic triggers.

[0063] In some embodiments, the subject is a human suffering from a disease associated with haemolysis. The disease may be selected from the group consisting of bacterial infections (e.g. Bartonella causing Carrion’s disease, Clostridium perfringens, enterohaemorrhagic strains of Escherichia coli, Shigella dysenteriae, Mycoplasma pneumoniae, Rickettsia rickeiisii. Salmonella enterica, Shiga-toxin producing strains of Escherichia coli, Streptococcus pneumoniae), viral infections (e.g. West Nile virus, Parvovirus Bl 9, Ebstein-Barr virus, human immunodeficiency virus, morbillivirus, paramyxoviruses, rubivirus, varicella-zoster virus, cytomegalovirus, hepatitis A virus, dengue virus, Flaviviridae, Filoviridae, Arenaviridae or Bunyaviridae), protozoan or parasitic infections (e.g. Plasmodium causing malaria, Trichomonas vaginalis causing trichomoniasis, Leishmania causing leishmanioses, Trypanosoma causing trypanosomal disease, Babesia causing babesiosis, Entamoeba histolytica), enzyme deficiencies (glucose 6-phosphate dehydrogenase deficiency [G6PDD], pyruvate kinase deficiency [PKD], glutathione synthetase deficiency [GSD]), inherited anaemias (thalassemia, sickle cell disease, hereditary spherocytosis, hereditary elliptocytosis), and adverse reactions (HELLP syndrome, thrombotic thrombocytopenic purpura (TTP) and haemolytic uremic syndrome (HUS), rhesus disease).

[0064] In some embodiments of the invention, the subject suffers from exertion -induced haemolysis due to strenuous exercise. The methods of the invention are particularly useful for these conditions as they are known to be associated with haemolysis and the diagnosis of haemolysis is therefore particularly useful.

[0065] The methods of the invention are further suitable for determining the presence or absence of haemolysis in a mammalian subject, comprising a step of detecting CAI in a urine sample from the subject, wherein the subject is a neonate who does not meet the threshold for phototherapy according to National Institute for Health and Care Excellence (NICE) guidelines. The criteria are set out in Table 1 below. This provides a major advantage in allowing neonates who otherwise would not have undergone tests for haemolysis to be easily and accurately diagnosed and treated accordingly. Without said method of the invention, the haemolysis would be left untreated, and symptoms of the underlying disease would worsen before the NICE criteria for phototherapy are reached. Therefore, the method of the invention enables neonates to be diagnosed and treated at a much earlier stage than current diagnostic methods. Depending on the deviation of the presence and / or concentration of the tested markers from the reference, the methods can determine the severity of haemolysis of a neonate. For example, where the deviation is large, a neonatal subject may suffer from a higher severity of haemolysis compared to a neonatal subject whose markers deviate less from the reference.

[0066] If the methods of the invention determine that haemolysis is present in a subject, then appropriate lifestyle changes can be instigated, such as withdrawing the putative trigger (e.g. food, drug, or activity), or medical attention can be sought.

[0067] The invention is also suitable for determining the presence or absence of an infective, immunological, pharmacological, biochemical or food or beverage trigger of haemolysis in a subject, comprising a step of detecting CAI in a urine sample from the subject. A trigger means a factor which causes haemolysis. For example, an infective trigger can be the presence of malaria or other disease which is known to cause haemolysis.

[0068] Foods and beverages that are known to trigger haemolysis in some subjects include, for example, blueberries, alcohol, legumes (such as fava beans, peanuts, peas and soybeans), as well as products with artificial dyes, containing menthol or animal, plant or fungi toxins or poisons (such as Acalypha indica poisoning. A pharmaceutical trigger may include medication such as some antibiotic (e.g. penicillins, cephalosporins, sulphonamides, rifampin), antiinflammatory (e.g. ibuprofen, paracetamol), anti-malarial (e.g. primaquine, quinine, quinidine), or chemotherapeutic (e.g. cisplatin, mitomycin) drugs. Toxins or poisons that are known to trigger haemolysis include Acalypha indica poisoning.

[0069] Carbonic anhydrase 1 (CAI)

[0070] The methods of the invention comprise a step of detecting or quantifying CAI, an enzyme belonging to the carbonic anhydrase enzyme family. Carbonic anhydrases (CAs) are zinc metalloenzymes that catalyse the hydration of carbon dioxide, and its reverse reaction. Therefore, CAs are involved in various biological processes including cellular respiration, calcification, acid-base balance, bone resorption and the formation of aqueous humour, cerebrospinal fluid, saliva and gastric acid.

[0071] Isoform CAI is a cytosolic protein that is most abundantly present in the cytosol of RBCs but also detected in the gastro-intestinal tract. The inventors have found that CAI can be used as a marker of haemolysis, due to its high presence in RBCs. Furthermore, CAI is small enough to pass the glomerular filter and is not recycled or reclaimed by tubular mechanisms, enabling detection in urine. The inventors have thus found that CAI is a good marker for detecting haemolysis in mammalian subjects.

[0072] There are a number of ways in which CAI can be detected, which are known to the skilled person, including ELISAs, western blots, immunohistochemistry, immunofluorescence, immunoprecipitation or lateral flow immunoassays.

[0073] The antibody may be polyclonal or recombinant but is preferably monoclonal. CAI may also be detected using antibody fragments, synthetic antibodies, binding peptides, aptamers, binders based on nucleic acids, small molecule binders or by direct detection.

[0074] Antibodies for detecting CAI are known in the art and include, for example EPR 193 and EPR23232-286 (Abeam), Carbonic Anhydrase I Antibody (200-1154-0100) (Thermo Fisher), and CAI antibody (AA 1-261).

[0075] CAI may also be detected by proteomic sequencing in the sample obtained from the mammalian subject. Suitable techniques for sequencing will be known to a skilled person. These include liquid chromatography-mass spectrometry, quadrupole mass analysis, nextgeneration sequencing, and others.

[0076] Haemoglobin

[0077] In some embodiments, the methods of the invention comprise a step of detecting or quantifying haemoglobin (Hb).

[0078] Haemoglobin is a protein found in RBCs. Normally, around a third of the total weight of RBCs is made of Hb. Hb contains iron(II) in its haem group that allows the protein to bind to molecular oxygen. Each haemoglobin tetramer molecule can bind up to four oxygen molecules.

[0079] Plasma free Hb is generally toxic and reclaimed by hepatic and macrophage mechanisms. Its large size, especially after binding to haptoglobin, hinders passage through the glomerular filter and any filtered protein is salvaged by processes downstream in the nephron because of the importance of reclaiming the body’s iron, over half of which is in RBCs. Therefore, presence of CAI but absence of Hb in urine samples is indicative of the release of ruptured RBC contents into the plasma (i.e. haemolysis inside blood vessels), whereas the presence of both Hb and CAI in a stoichiometric ratio would indicate bleeding along the urogenital tract, a contamination of the urine sample. Assays for measuring Hb levels are well known in the art, for example through detection with a suitable antibody or from its light-absorbance properties.

[0080] The subject

[0081] The methods of the invention are for determining the presence or absence of haemolysis in a mammalian subject, comprising a step of detecting CAI in a urine sample from the subject. In some embodiments the subject is not a human neonate.

[0082] The terms “subject,” “individual,” and “patient” are used interchangeably herein to refer to a mammal.

[0083] The methods of the invention will be particularly suitable where the subject has a high level of physical activity as exercise (in particular, strenuous exercise) has been associated with increased levels of haemolysis due to mechanical damage.

[0084] In some embodiments, the subject is a human, a dog, a cat or a horse. The horse may be a racehorse as these undergo an intense training regime, can present with haemolytic risks due to genetic traits acquired through selective breeding, and haemolysis can negatively affect performance. The testing of dogs is also particularly useful as many dogs are inbred and it is not always feasible or economical to undertake a full blood count in a clinical setting.

[0085] Where the mammal is a human, the human may be a child or an adult. The human may be at least 18 years old, at least 17 years old, at least 16 years old, at least 15 years old, at least 14 years old, at least 13 years old, at least 12 years old, at least 11 years old, at least 10 years old, at least 9 years old, at least 8 years old, at least 7 years old, at least 6 years old, at least 5 years old, at least 4 years old, at least 3 years old, at least 2 years old, at least 1 year old, at least 6 months old, at least 3 months old, at least 29 days old. In some embodiments, the subject is a human who is at least 29 days old, at least 1 year old, at least 5 years old, at least 10 years old, at least 15 years old or at least 18 years old. Preferably, the human is at least 29 days old.

[0086] A further method of the invention determines the presence or absence of haemolysis in a mammalian subject, comprising a step of detecting CAI in a urine sample from the subject, wherein the subject is a neonate who does not meet the threshold for phototherapy according to the NICE guidelines. The term “neonate” refers to a child of 28 days of age or younger.

[0087] Newborn babies are known to be at high risk of haemolysis due to birth complications such as immune reactions (ABO and Rhesus incompatibility) and infections (e.g. neonatal sepsis). As such, testing on neonates must specifically distinguish haemolytic crises from milder forms arising from the replacement of foetal Hb with the adult variant (O'Brien R.T. and Pearson H.A., 1971, JPediatr, 79(1), pp. 132-138.). In newborns, the diagnostic power of Hb, bilirubin or Hp can be compromised by hepatic immaturity, which could lead to erroneous readings (Carter K. and Worwood M., 2007, IntJ Lab Hematol, 29(2), pp. 92-110; Naryzny S.N. and Legina O.K., 2021, Biochem Mose Suppl B Biomed Chem, 15(3): 184-198.). Certain RBC- related genetic traits predispose to haemolysis, including disorders of Hb (e.g. sickle cell disease (Kato G.J. et al., 2017, J Clin Invest, pp. 750-760)), of the cytoskeleton (e.g. hereditary spherocytosis) or of enzymes (e.g. glucose 6-phosphate dehydrogenase deficiency, G6PDD). In G6PDD, glutathione production in RBCs is compromised, rendering RBCs susceptible to rupture upon oxidative stress, such as that arising from ingestion of some drugs or foods (Morelli A. et al., 1987, Blood, 69(6), pp.1753-1758.).

[0088] A problem with the present testing methods in neonates is that the presence of a critical level of haemolysis (requiring phototherapy) is inferred from surrogates such as bilirubin, in accordance with NICE or similar international guidelines. Serum or transcutaneous bilirubin is, however, indirectly associated with haemolysis and can produce false readings, for example due false - negative due to hepatic immaturity or false positive in Crigler-Najjar syndrome. Accordingly, neonates may be incorrectly assessed as not haemolysing when, indeed, treatment is required. This drawback is overcome by the present invention which provides a more direct and earlier determination of haemolysis.

[0089] A neonate will be considered as not meeting the threshold for phototherapy according to the NICE guidelines if the total serum bilirubin levels are below the concentration shown in Table 1:

[0090] Such infants will be particularly suitable as subjects in the methods of the invention as critical levels of haemolysis may easily be missed if they are assessed according to their bilirubin levels alone. Malaria

[0091] A further group who is at particular risk of haemolysis are subjects suffering from infections, including bacterial infections (e.g. sepsis (Effenberger-Neidnicht K. and Hartmann M. 2018, Inflammation, 41(5), pp. 1569-1581)) or protozoan infections, notably malaria where parasites invading RBCs can cause rupture (Poespoprodjo J.R. et al., 2023, Lancet, 402(10419), pp. 2328-2345).

[0092] Haemolysis in malaria is exacerbated by adverse reactions to medications, most notably post- artesunate delayed haemolysis (Jaureguiberry S. et al., 2014, Blood, 124(2), pp. 167-175). As many as 5% of cases of microangiopathic haemolytic anaemia are drug -induced, a level that should mandate proactive screening. Collectively, these triggers and genetic predispositions to their adverse actions are present globally, but tend to be more prevalent in developing regions (Murray N.A. and Roberts I.A., 2007, Arch Dis Child Fetal Neonatal Ed, 92(2):F83-8.; Slusher T.M. et al., 2017, BMJ Paediatr Open, 1 (l):e000105; Mitra S. and Rennie J., 2017, Br J Hosp Med (Lond), 78(12), pp. 699-704) and associated with certain ethnicities (e.g. sickle cell trait, Rhesus disease, G6PDD) (Dhaliwal G. et al., 2004, American family physician, 69(11), pp.2599-2606; Moise K.J., 2013, Clinical advances in hematology & oncology : H&O, 11(10), pp. 664-666.; Kumar S. and Regan F., 2005, Bmj, 330(7502), pp. 1255-1258.; Moise K.J., 2008, Obstetrics and gynecology, 112(1), pp. 164-76), thus costs are important to factor for equitable access to testing.

[0093] Malaria is a disease caused by Plasmodium parasites, which are transmitted into humans by female mosquitos. The parasites infect RBCs and can cause their rupture.

[0094] Malaria causes a number of symptoms including fever, fatigue, vomiting and headaches. In severe cases symptoms can include jaundice, coma and even death. Symptoms usually begin approximately 10-15 days after infection, and can be recurrent for months if not properly treated.

[0095] Malaria is a widespread disease in tropical and subtropical environments around the equator. In 2022, 249 million cases of malaria worldwide resulted in approximately 608,000 deaths. Malaria is commonly associated with poverty and has negative effects on the economic development of affected countries. The invention allows for the determination of the presence or absence of a haemolytic reaction to malaria parasite infection or an adverse reaction to anti-malarial drugs used in a non-invasive sample from a subject, comprising a step of detecting carbonic anhydrase 1 (CAI).

[0096] The invention further allows for determining the severity of haemolysis.

[0097] The methods of the invention can also be used to monitor the treatment efficacy, safety of a drug treatment and dose-escalation in a patient. In particular, some drugs used to treat malaria patients have been known to cause haemolysis. By detecting and / or measuring CAI as described herein, the methods of the invention allow a skilled person to determine the efficacy, safety and / or dose-escalation of a drug treatment. In turn this permits the medical professional to make an informed therapy choice and determine accurate doses, for example in malaria patients. This is particularly important in malaria patients because malaria treatment can exacerbate haemolysis in some patients. This may include managing dosage of hydroxyurea in sickle cell patients, which is expected to reduce haemolysis if correctly titrated; managing drugs that carry a haemolytic risk such as antibiotics (cephalosporins, penicillins), quinidine, antimalatials (primaquine, chloroquine), sulfa drugs. The methods of the invention allow the detection of haemolysis in a cost-effective, convenient and accurate manner and so the methods of the invention are particularly suitable for monitoring treatment in malaria patients.

[0098] In this aspect of the invention, CAI is preferably quantified and may further be compared to a reference value. If the CAI concentration is equal to or lower than the reference value no haemolysis is present and the drug treatment is deemed to be efficacious and / or safe. Conversely, if the CA concentration is higher than the reference value, haemolysis is deemed to be present and the treatment can be considered not efficacious and / or unsafe.

[0099] Marker detection and quantification

[0100] A skilled person may use any suitable technique known in the art to detect or quantify CAI alone or in combination with other marker(s) in a method according to the invention. In some embodiments, the markers are CAI and Hb.

[0101] The marker(s) may be detected or quantified by interaction with a ligand or ligands, 1 -D or 2-D gel-based analysis systems, liquid chromatography, combined liquid chromatography and any mass spectrometry techniques including MSMS, ICAT(R) or iTRAQ(R), agglutination tests, thin-layer chromatography, NMR spectroscopy, sandwich immunoassays, enzyme linked immunosorbent assays (ELISAs), radioimmunoassays (RAI), enzyme immunoassays (EIA), lateral flow immunoassays / immunochromatographic strip tests, western blotting, immunoprecipitation, particle -based immunoassays including using gold, silver, or latex particles and magnetic particles or Q-dots, recombinant antibodies and their genetically engineered forms, binding peptides, aptamers (or binders based on nucleic acids), small molecule fluorescent binders or any other suitable technique known in the art.

[0102] The methods of the invention are preferably practised using a lateral flow immunoassay. This is preferred because the inventors have found that these assays provide accurate results in the context of the invention. Furthermore, these assays are easy to use and do not require specialised equipment, making them suitable for home testing.

[0103] In preferred embodiments, the markers are detected using a lateral flow assay device. A rapid lateral flow assay device generally consists of a system of overlapping porous materials containing the dried components needed to perform the test. These membranes are assembled in small strips, which may be placed into a plastic housing for ease in handling. The lateral flow assay device may comprise (i) a sample receiving region; and (ii) a capture membrane positioned downstream of the sample receiving region and comprising a test region including an antibody suitable for detecting CAI. It preferably further comprises an antibody suitable for detecting Hb. Suitably, the device may further comprise an antibody for detecting a positive control and / or a reference. The device may further comprise an absorbent pad (arranged downstream), also known as a wicking or sink pad, arranged to collect the fluid flowing through the test system and prevent any backflow of fluid.

[0104] The capture membrane typically comprises or consists of nitrocellulose and may optionally further contain fibres coated with a surfactant. It may also comprise particles of a polymer, such as polyethylene, fused together. Such particles can be spherical particles. An example of this type of element is the POREX Lateral-Flo element (POREX Corporation, Fairbum, Ga.).

[0105] The capture membrane comprises one or more antibodies for detecting CAI either alone or in combination with Hb. The one or more antibodies are typically labelled with a suitable detection reagent. Such reagents are known to a skilled person and include, for example, a coloured material, a fluorescent label, or a chemiluminescent label, a redox label such as ferrocyanide, a radioactive label, a radiofrequency label, an enzymatic label, and / or a bioluminescent label. Where more than one marker is detected a combination of labels may be used. For example, the assay may use fluorescent materials that fluoresce at different wavelengths.

[0106] The label may also be a particle. Such particles can be colloidal gold particles, colloidal sulphur particles, colloidal selenium particles, colloidal barium sulphate particles, colloidal iron sulphate particles, colloidal metal sulphide particles, colloidal lead selenide particles, colloidal cadmium selenide particles, colloidal metal iodate particles, colloidal metal phosphate particles, colloidal metal ferrite particles, colloidal silver halide particles, colloidal silica particles, colloidal carbon or colloidal metal (hydrous) oxide particles.

[0107] The lateral flow test may be qualitative or it may be quantitative. Quantitative lateral flow immunoassays are known in the art and have been described for example in W02005 / 117556.

[0108] In some embodiments of the methods of the invention, the step of detecting or quantifying detect CAI or Hb comprises performing an ELISA assay. In a further embodiment, the ELISA assay is a sandwich ELISA assay. A sandwich ELISA assay comprises steps of capturing the marker(s) to be detected (CAI or Hb) or whose concentration is to be determined using a "capture antibody” already bound to a plate, and detecting how much of the marker(s) has been captured using a "detection antibody”. The detection antibody may be pre-conjugated to a label such as the enzyme HRP (Horse Radish Peroxidase). The ELISA plate may then be exposed to the labelled detection antibody, such that the labelled detection antibody binds to the captured marker(s). After exposure to the labelled detection antibody, the ELISA plate should then be washed to remove any excess unbound labelled detection antibody. The washed plate can then be exposed to an agent whose properties are changed by the label (of the detection antibody) in a measurable manner. The concentration of the detection antibody may then be determined. For example, if the detection antibody is labelled by e.g. conjugation to HRP, the ELISA plate may be exposed to 3,3',5,5'-Tetramethylbenzidine (TMB) substrate. The concentration of the detection antibody, and therefore the concentration of the marker(s) in the original non-invasive sample, may then be determined by quantitation of the colour change corresponding to the conversion of TMB into a coloured product.

[0109] The ELISA assay may be a qualitative ELISA assay. In the context of the invention, a qualitative ELISA assay is one that is performed without determining a numerical value for the concentration of the marker(s) in the non-invasive sample and / or without determining, or using a previously determined, numerical value for the concentration of the marker(s) as a reference. For example, in a qualitative ELISA assay, the intensity of a coloured product (e.g. a coloured product produced by an enzyme linked to a detection antibody) may be compared to a reference. In this embodiment, the intensity of the coloured product is indicative of the concentration of the marker(s) without the need to determine a numerical value for the concentration of the marker(s). The reference used in a qualitative ELISA of this type may be a previously determined threshold intensity of the coloured product.

[0110] The methods of the invention preferably comprise a step of comparing the concentration of the one or more marker(s) to a reference. Suitably, the reference may be a concentration of the one or more marker(s) determined for a sample obtained from one or more healthy individual(s). A healthy individual in this context is a subject who is not affected by haemolysis. The sample may also be compared to a positive control which is a sample from an individual who is known to suffer from haemolysis.

[0111] CAI is not normally present in the urine of healthy individuals. Accordingly, the concentration in the reference is likely to be close to 0. A subject will thus be considered to suffer from haemolysis if the concentration of CAI is >4ng / mL, >5ng / mL, >6ng / mL, >7ng / mL, >9ng / mL, or >10ng / mL.

[0112] For the one or more marker(s), where the sample is compared to a positive control, the concentration will be similar to the reference value. Similar in this context may mean that the value is with a range of ± 5%.

[0113] The reference used in the methods of the invention may be an average concentration of the marker(s) determined for multiple samples obtained from a single healthy individual.

[0114] Alternatively, the reference may be an average concentration of the one or more marker(s) determined for multiple samples prepared from multiple individuals. In this context, “one or more samples” may be at least 10, 100, 500, 1000, 10,000, 100,000 or 1,000,000 samples.

[0115] In another embodiment, the reference can be an average concentration of the one or more marker(s) previously determined for one or more samples prepared from a healthy individual or a subject suffering from haemolysis. In such embodiments, a numeric comparison may be made by comparing the concentration of the one or more marker(s) determined for the sample obtained in the invention to the reference. The advantage of this is not having to duplicate the analysis by determining a reference in parallel each time a sample from a subject is analysed.

[0116] Suitably, the reference may be matched to the subject being analysed e.g. by species and by age. For example, the reference may suitably be matched to specific patient sub-groups e.g. younger subjects.

[0117] In some embodiments, the concentration of the one or more marker(s) determined may be compared to a concentration of the one or more marker(s) previously determined from a non- invasive sample obtained from the same subject. In these embodiments, the previously determined concentration of the one or more marker(s) is used as the reference. This can be beneficial in monitoring the subject. In some embodiments, the non -invasive sample is urine.

[0118] In some embodiments, haemolysis is less likely to be present if the concentration of the one or more marker(s), analysed according to the methods of the invention, is normal compared to the reference. The concentration of the one or more marker(s) may be said to be normal compared to the reference if there is no statistically significant difference between the concentration of the one or more marker(s) and the reference. For example, the concentration of the one or more marker(s) may be said to be normal compared to the reference if the difference between the concentration of the one or more marker(s) and the reference is less than two, or less than one standard deviations.

[0119] In some embodiments, the subject may be diagnosed as having haemolysis if the concentration of the one or more marker(s) determined for the sample is higher than the mean value determined for healthy individuals. In such embodiments, the "mean value determined for healthy individuals" is a type dereference” as defined herein. It should be understood that the various embodiments and characteristics of a "reference" as defined herein may also apply to the "mean value determined for healthy individuals” . In some embodiments, the concentration of the one or more marker(s) is higher than the mean value determined for healthy individuals if it is higher than the mean value determined for healthy individuals plus a multiple of the standard deviation of the mean value determined for healthy individuals, for example, higher than the mean value plus one, two, three, four, or five standard deviations of the mean value determined for healthy individuals.

[0120] In some embodiments, the reference value for CAI ranges from 0 ng / mL - 4 ng / mL. In some embodiments, absence of CAI indicates no haemolysis. An increased concentration when compared to the reference value indicates haemolysis. In some embodiments, a CAI concentration from 0 ng / mL - 4 ng / mL indicates physiological (benign) haemolysis. In some embodiments, a concentration of CAI >4ng / mL indicates pathological (severe) haemolysis.

[0121] In some embodiments, the reference value for Hb is 0 mg / mL. In some embodiments, an increased concentration when compared to the reference value indicates haemolysis. In some embodiments, a value of greater than 0 mg / mL indicates haemolysis.

[0122] In some embodiments, a positive CAI result and a positive Hb result indicates contamination of the sample from bleeding along the urogenital tract, or kidney failure, rather than haemolysis inside blood vessels. In some embodiments, a positive CAI result and a negative Hb result indicates intravascular haemolysis. In some embodiments, a positive CAI result and a negative Hb result indicates extravascular haemolysis that has released red cell contents into the blood stream. A positive CAI result, a negative Hb and increased bilirubin may indicate extravascular haemolysis. A positive CAI result, a negative Hb and increased free plasma haemoglobin may indicate intravascular hemolysis. Sensitivity and Specificity

[0123] The methods of the invention allow haemolysis taking place inside blood vessels to be detected with high specificity and high sensitivity. In the field of medical diagnostics and as used herein the term “sensitivity” (also referred to as the true positive rate) refers to a measure of the proportion of actual positives that are correctly identified as such. In other words, the sensitivity of a diagnostic test may be expressed as the number of true positives i.e. individuals correctly identified as having a disease as a proportion of all the individuals having the disease in the test population (i.e. the sum of true positive and false negative outcomes). Thus, a high sensitivity diagnostic test is desirable as it rarely misidentifies individuals having the disease. This means that a negative result obtained by a highly sensitive test has a high likelihood of ruling out the disease.

[0124] In the field of medical diagnostics, and as used herein, the term “specificity” (also referred to as the true negative rate) refers to a measure of the proportion of actual negatives that are correctly identified as such. In other words, the specificity of a diagnostic test may be expressed as the number of true negatives (i.e. healthy individuals correctly identified as not having a disease) as a proportion of all the healthy individuals in the test population (i.e. the sum of true negative and false positive outcomes). Thus, a high specificity diagnostic test is desirable as it rarely misidentifies healthy individuals.

[0125] In the field of medical diagnostics, and as used herein, the term “positive predictive value (PPV)” refers to the proportion of all positive outcomes generated by diagnostic test that are true positive outcomes. Put another way, PPV can be defined as the number of true positive outcomes divided by the sum of true positive outcomes and false positive outcomes. In the field of medical diagnostics, and as used herein, the term “negative predictive value (NPV)” refers to the proportion of all negative outcomes generated by diagnostic test that are true negative outcomes. Put another way, NPV can be defined as the number of true negative outcomes divided by the sum of true negative outcomes and false negative outcomes.

[0126] In the field of medical diagnostics, and as used herein, a “Receiver Operating Characteristic (ROC) curve” refers to a plot of true positive rate (sensitivity) against the false positive rate (1 - specificity) for all possible cut-off values. In the field of medical diagnostics, and as used herein, a “Y oudens index” refers to the cut-off point at which the distance between the ROC curve and the line of chance (45-degree line) is highest. These terms are well known in the art and to the skilled person. The specificity and / or sensitivity of a method may be determined by performing said method on samples which are known to be positive samples (e.g. samples from patients having a gynaecological cancer) and / or samples which are known to be negative samples (e.g. samples from healthy individuals). The extent to which the method correctly identifies the known positive samples (i.e. the sensitivity / true positive rate of the method) and / or the known negative samples (i.e. the specificity / true negative rate of the method) can thus be determined.

[0127] In some embodiments of the methods of the invention, the method has a sensitivity for detecting haemolysis of at least about 40 %, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or 99%.

[0128] In some embodiments of the methods of the invention, the method has a sensitivity for detecting haemolysis of at least about 50 %, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or 99%.

[0129] Samples

[0130] The methods of the invention comprise measuring the markers in a non -invasive sample. As used herein, the term “non-invasive” refers to a step that does not require piercing of the skin, such as that required to obtain a tissue sample in order to perform biopsy.

[0131] The sample is preferably urine. This is preferred because urine is easy to obtain using non- invasive techniques for point-of-care and minimal-laboratory use whilst still providing an accurate detection of haemolysis in accordance with the invention.

[0132] The invention may also be practised with other samples, such as blood, serum or plasma.

[0133] The methods of the invention may comprise a step of obtaining the urine sample from the subject.

[0134] The methods of the invention may use the entirety of a sample from a subject. It may also use a part of the sample for testing. For example, the part of the sample used for testing may have a volume of less than 150pl, less than 200pl, less than 250pl, less than 500pl, less than 750pl, less than 1ml or less than 2ml. In some embodiments it has a volume of less than 1ml.

[0135] The methods of the invention can be practised using a drop from the sample. The part of the sample that is tested may thus have a volume of more than about lOOpl, 200pl, 250pl, 300pl, or 400pl. It may have a volume of between lOOpl - lOOOpl, lOOpl - 900pl, lOOpl - 800pl, lOOpl - 700pl, lOOpl - 600pl, lOOpl - 500pl, lOOpl - 400pl, or lOOpl - 300pl.

[0136] The sample may be collected using any clean, preferably sterile, container. Examples are neonatal urine bags, cotton balls, 1.5ml screw cap tubes, 15 / 50ml conical tubes. In some embodiments, the subject is neonate, wherein the neonate does not meet the threshold for phototherapy according to the NICE guidelines. Samples may be taken over the first ten days of life. Samples may be taken on day 1, day 2, day 3, day 4, day 5, day 6, day 7, day 8, day 9 or day 10 of postnatal life.

[0137] A urine sample can be obtained using any technique known in the art that the skilled person would be aware of. For example, a urine sample may be obtained from a subject by asking the subject to urinate into a collection vessel, such as a sterile container.

[0138] In some embodiments, the methods of the invention do not involve a separate step of collecting a sample. For example, a suitable test (such as a lateral flow immunoassay) can be included into a diaper which allows for the methods of the invention to be performed in an easy and convenient manner, in particular where the subject is a child or an elderly bedbound person.

[0139] As used herein, the term “urine sample” encompasses any sample derived from the urine sample originally obtained from the subject. For example, the urine sample may be processed by centrifugation to obtain supernatant used in the method of the invention.

[0140] In some embodiments the methods of the invention may comprise a step of obtaining a non- invasive sample from a subject.

[0141] The sample may be analysed immediately after collection. It may also be stored prior to analysis, for example at 4°C, -20 °C or -80 °C. In some embodiments the sample may be mixed with a buffer, for example in order to increase the stability of the sample. Suitable buffers will be known to a skilled person and include phospho-buffered saline (PBS), inhibitors of protease enzymes and stabilizing blocking buffers (e.g. PBS or borate buffer containing BSA, casein, sucrose, polyvinyl alcohol, detergents like Tween20, SDS or Brij-35).

[0142] Kits

[0143] The invention also provides a kit suitable for use in the methods of the invention. The kit may contain the reagents necessary to perform the methods of the invention. For example, it may contain one, two or more antibodies, wherein at least one antibody can detect CAI and optionally at least one antibody can detect Hb. The antibodies are preferably monoclonal or recombinant antibodies.

[0144] In a preferred embodiment the kit comprises a lateral flow immunoassay.

[0145] A kit according to the invention may further comprise reagents, such as buffers suitable for carrying out the methods of the invention. The kit may also contain receptacles suitable for collecting the non-invasive sample. For example, where the sample is urine, the kit may contain a collection vessel such as a sterile container, neonatal urine bags, cotton balls, screw cap tubes.

[0146] Uses

[0147] Further aspects of the invention relate to the use of the kits or lateral flow devices of the invention in a method of determining haemolysis inside blood vessels in a subject. The uses and methods of the invention may be in vitro or ex vivo.

[0148] Other

[0149] It is to be understood that the different embodiments disclosed may be tailored to the specific needs in the art. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments of the invention only, and is not intended to be limiting.

[0150] The singular forms “a”, “an”, and “the” include plural references unless the content clearly dictates otherwise.

[0151] Furthermore, when referring to “>x” herein, this means equal to or greater than x. When referred to “<x” herein, this means less than or equal to x.

[0152] The term “comprises” (comprise, comprising) should be understood to have its normal meaning in the art, i.e. that the stated feature or group of features is included, but that the term does not exclude any other stated feature or group of features from also being present. For example, a lysis buffer comprising a detergent may contain other components.

[0153] The term “consists of should also be understood to have its normal meaning in the art, i.e. that the stated feature or group of features is included, to the exclusion of further features. For example a lysis buffer consisting of a detergent contains detergent and no other components. A lysis buffer comprising a detergent consisting of polysorbate 80 may comprise components other than detergents but the only detergent in the lysis buffer is polysorbate 80.

[0154] For every embodiment in which “comprises” or “comprising” is used, we anticipate a further embodiment in which “consists of’ or “consisting of is used. Thus, every disclosure of “comprises” should be considered to be a disclosure of “consists of.

[0155] Examples

[0156] Example 1 - Sampling urine from participants presenting a range of haemolytic risk

[0157] URICA2 (Urinary excretion of Carbonic Anhydrase Study 2) recruited 215 participants in

[0158] Oxford and London in the UK, Chattogram and Ramu in Bangladesh, and four health centres in

[0159] TJ Peru’s Loreto region (Figure 1A). Reference control urine samples were collected from 56 adult volunteers in good self-reported health, 27 of whom were female in Oxford. These individuals did not provide blood samples for additional measurements and were assumed to fall within reference ranges.

[0160] To assess urinary CAI excretion across a spectrum of haemolytic severity, urine samples were collected from patients with sickle cell disease (SCD), a condition characterized by intravascular haemolysis, as well as from individuals with conditions less commonly associated with intravascular haemolysis (9 females, 9 males). Markers of haemolysis were measured by standard blood tests.

[0161] Urine samples from the Bangladeshi cohort (34 males, 16 females, 5 sex unknown) were sampled from children (10 years or older) and adults, including 30 complicated and 20 uncomplicated cases of falciparum malaria, alongside 5 healthy controls (Figure IB). Twentyeight patients with complicated and 3 patients with uncomplicated cases provided urine samples through a Foley catheter, an intervention that may cause urogenital bleeding and potentially contaminate samples with CAI released in situ, rather than from an intravascular event. This cohort is suitable for testing the specificity of urinary CAI excretion as a marker of intravascular haemolysis. Blood in urine was assessed by urinalysis and haemolysis scoring (Figure 1C).

[0162] To evaluate the diagnostic value of urinary CAI in a clinically more diverse group, urine samples were obtained from 72 adults (40 males, 32 females) attending health centres in the Loreto region of the Peruvian Amazon where malaria is endemic (Figure ID). Microscopy indicated malaria in 7 females and 15 males, typically caused by Plasmodium vivax, a parasite infecting immature RBCs and causing less severe haemolysis.

[0163] To study the evolution of CAI excretion in early postnatal life, 32 newborns were recruited among admissions to neonatal intensive care of London Evelina’s Children Hospital. Participants presented a range of medical conditions and a spectrum of haemolytic severity, from mild (physiological) to critical. Up to daily urine samples were collected over the first 10 postnatal days (median: 6 samples / participant), alongside blood sampling (full blood count, serum bilirubin, Hb, CRP), skin bilirubinometry, clinical diagnosis and DAT result (Figure IE).

[0164] Analysis of the patient populations allowed the definition of threshold values as shown in the

[0165] Table 2:

[0166] Example 2 - Urine samples span a range of CAI immunoreactivity

[0167] Urine was tested for Hba and CAI immunoreactivity by western blot (using albumin as a loading control) and ELISA. Haemolysis is expected to raise CAI signal; when this occurs with little or no rise in Hb, a haemolytic event inside blood vessels is inferred on the basis that glomerular fdtration favours passage of CAI over Hb (Figure 2A). Conversely, a significant and stoichiometric rise in Hb immunoreactivity would indicate rupture of RBCs in blood- contaminated urine (e.g., urogenital bleeding or menstruation), potentially leading to a falsepositive diagnosis of haemolysis inside blood vessels. To assess this concern, Hb and CAI levels were compared after calibrating against blood lysates (1: 10,000 v / v dilution) which release Hb and CAI stoichiometrically.

[0168] The quantitative power of calibrated ELISA measurements was exploited to relate CAI with Hb signal and to determine thresholds for positivity. Analysis of reference controls (Oxford) by Grubb’s test identified eight outliers, thus defining a CAI positivity threshold of 0. 142. A similar process for Hb identified two outliers and defined a threshold of 0.193 (Figure 2B). Applying these thresholds identified urines from non -haemolyzing cases (CA1- / Hb-). A complete separation of CAI from Hb in CAHTHb- urines was considered highly likely to arise from haemolysis inside blood vessels. Cases where the blood-calibrated CAI signal exceeded the Hb signal by at least 33% were deemed to have a component due to haemolysis inside blood vessels, based on analysis of RBC lysates at various dilutions. Urine samples with comparable Hb and CAI positivity were classified as ambiguous due to blood contamination masking any true intravascular or extravascular haemolytic event.

[0169] Among samples obtained in Oxford, the reference cohort excreted mostly CA1- / Hb- urines (84%), whereas hospital patients produced Hb-negative urines spanning a range of CAI positivity. In the Bangladeshi group, 31 malaria cases (55%) had at least a component of haemolysis, 9 urines (16%) were deemed significantly contaminated by bleeding, and healthy controls were CA1- / Hb-. In the Peruvian cohort, 40 (56%) samples were classified as haemolysis, two (3%) contaminated by bleeding, and the remainder CA1- / Hb-. Of 175 samples taken from 32 newborns, 108 (62%) were CAI -positive, of which only three were also Hb- positive but only one classified as blood-contaminated. Classification of participants by CAl / Hb status is summarized in Figure 2C / D.

[0170] Western blots were quantified by densitometry and scored visually on a scale zero (unambiguously negative), one (faint signal), two (strong signal) and three (very strong signal), relative to positive controls (Figure 2E). For technical repeats, the average was rounded up. Reference controls were mostly CAI and Hb negative, with occasional low -level CAI detection (score 1). In contrast, among hospital admissions in Oxford, malaria-infected participants in Bangladesh, health clinic attendees Peru, and neonates in intensive care in London excreted a range of CAI in urine. Among the examples shown, most CAl-positive urines were not associated with proportional Hb-positivity, which contrasts with the stoichiometric coupling measured with urines spiked with RBC lysates. Time courses from four newborns were selected to illustrate an unstable CAI time course, a low-amplitude early CAI transient, and two cases of sustained and elevated urinary CAI excretion. Western blot densitometry (Figure 2F) and ELISA absorbance measurements (Figure 2G) correlated well with western blot score, noting that some variation is expected because of differences in the methods and readout non-linearity.

[0171] Example 3 - Haematological conditions characterized by intravascular haemolysis result in urinary CAI

[0172] Participants recruited to the study included ten sickle cell disease (SCD) patients, and patients who are not expected to be haemolyzing intravascularly (hereditary spherocytosis, hemochromatosis, Diamond Blackfan anaemia). Urines were tested for CAI by ELISA (Figure 3A-B) and western blot (Figure 3C). SCD patients were anaemic with raised reticulocyte counts, plasma LDH and serum bilirubin, consistent with intravascular haemolysis (Figure 3D- G). Compared to reference controls and patients unlikely to be haemolyzing, SCD participants excreted significant CAI levels, detected by western blot and ELISA. Leveraging the quantitative power of ELISA, correlations were noted against Hb, bilirubin, LDH and reticulocyte count (Figure 3H-L). Critically, no urine sample was Hb-positive, indicating separation of CAI from Hb, as expected from RBCs rupture proximally to glomerular filtration. Linear mixed effect modelling produced a strong correlation between a function of LDH and Hb with urinary CAI ELISA (Figure 3M). Example 4 - Raised plasma free haemoglobin is associated with urinary CAI excretion

[0173] The Bangladeshi cohort spans a wide range of haemolytic states for seeking correlations between urinary CAI and markers of haemolysis (Figure 4A). Complicated malaria cases had reduced blood Hb, higher PFH (mostly >0.05 g / L), LDH (>250 U / mL), and elevated bilirubin. Raised creatinine (>3 mg / dL) was detected in 30% of complicated cases and indicated renal failure. Using a 2.2 mg / dL cut-off, the complicated malaria group could be sub-divided into a low and high creatinine group (CACrL, CACrH). CAI ELISA and western blot score revealed a trend from negative in healthy controls, intermediate in uncomplicated (U), to substantial in CACrL (Figure 4B-D). CAI scores for the CACrH group were more variable, which may reflect an influence of urinary output on the relationship between CAI excretion and haemolysis. Similar trends were noted for urinary CAI quantified by ELISA, with some of the strongest CAI signals associated with urinary Hb, raising the possibility of blood contamination. Indeed, higher scores for blood on urinalysis test associated with higher CAI signals (Figure 4B); therefore, urine CAI levels in at least some urines may reflect erythocyturia, rather than haemolysis inside blood vessels.

[0174] After removing cases of suspected renal failure, serum LDH correlated strongly with CAI ELISA (Figure 4E). Similarly, after removing cases of suspected renal failure, plasma free Hb (PFH) correlated strongly with CAI ELISA (Figure 4F). Further analysis of all samples by decision tree ranked predictors of urinary CAI western blot score (Figure 4G), identifying PFH as the strongest predictor overall. Regression analysis identified predictors of CAI ELISA (Figure 4H). Additional predictors were included in a regression model (Figure 41), and logtransform was applied to PFH to account for spread. This model produced an excellent fit, without overfitting or collinearity (variance inflation factor <1.5) using PFH, urine color (normal / abnormal), urine protein (score 0-3), and interactions thereof. Thus, urinary CAI excretion was significantly correlated with PFH, a direct clinical marker of haemolysis.

[0175] The Bangladeshi cohort was sub-divided by thresholding PFH at 0.07 g / L (Figure 4J), also show by patient breakdown (Figure 4K). Subthreshold samples of normal urine color were mostly CAI -negative (“Group- 1”) whereas subthreshold samples of abnormal color were generally CAI -positive and in excess of urinary Hb, thus determined to represent haemolysis inside blood vessels that evades detection on PFH (“Group-2”). Participants with suprathreshold PFH generally excreted CAl-positive urine (“Group-3”). Among CAl-negative samples, most (13 / 15) were from participants with low PFH (<0.07 g / L), consistent with a non-haemolyzing group that included healthy, complicated and uncomplicated malaria cases. Among CA1- positive urines, 17 out of 31 were collected from mostly complicated malaria cases and defined as haemolyzing based on raised PFH. The remaining 14 samples, principally from uncomplicated malaria, were deemed to indicate haemolysis inside blood vessels but below the detection threshold of PFH measurements. Overall, CAI WB score was sensitive in detecting cases of raised PFH (sensitivity 93%; Figure 4K). CAI testing was able to detect a haemolytic event in cases that are subthreshold for PFH (Figure 4K)

[0176] Example 5 - Urinary CAI excretion correlates with inflammatory markers

[0177] The Peruvian cohort included participants with laboratory diagnoses of malaria (mostly Plasmodium vivax), non-malarial febrile cases, as well as participants with no evidence of infection. Urine samples were paired with measurements of blood Hb, direct and indirect bilirubin, Hp, and CRP to test the relationship between markers of disease (e.g., inflammation) and urinary CAI excretion (Figure 5A). Samples were annotated with information on the presence of fever, malaria test, and anti-malarial prescriptions (chloroquine with primaquine). Malaria diagnosis was significantly associated with reduced Hp and higher direct bilirubin, as expected from haemolysis, although the effect sizes were small, with most samples remaining within their reference ranges. In contrast, CRP was substantially raised among malaria-positive diagnoses, with all exceeding the threshold of 3 mg / L, although a few non-malaria cases had high CRP (Figure 5B). Febrile participants included both malaria-positive and malaria-negative, with low or high CRP. Thus, the cohort comprises various causes of raised CRP, providing an opportunity to relate urinary CAI excretion with this marker of inflammation.

[0178] Decision tree analysis against urinary CAI western blot score indicated CRP levels as having the strongest influence (Figure 5C). There was no significant association between CAI score and age or sex, but a positive correlation to malaria diagnosis and malaria medication (Figure 5D). Of the 22 malaria-positive cases, 18 produced CAI immunoreactivity on western blot. A lack of correlation between urinary CAI and Hp corroborates its weak relationship with malaria status. A positive correlation to CAI score emerged for direct bilirubin and CRP (Figure 5E), the latter consistent with decision tree analysis. Thresholding CRP at 22 mg / L separated the cohort into a CAllow (score 0 or 1) and CAlhigh (score 2 or 3) group, with specificity and sensitivity >90% (Figure 5F / G). CRPhigh / CAlhigh participants were typically malaria-positive, whereas CRPlow / CAllow cases were either a nominally control group or non-malarial febrile. The cohort can be further stratified by CRP status (<3 mg / L, 3-22 mg / L, >22 mg / L), showing near-complete ELISA CAI positivity for the highest inflammatory status (Figure 5H). Overall, significant urinary CAI excretion (score 2-3) was a sensitive and specific surrogate of inflammatory states. Moreover, urinary CAI status was superior to Hp or bilirubin in separating patients by disease severity and, ostensibly, haemolytic risk.

[0179] Example 6 - Urinary CAI stratifies newborns by haemolytic severity

[0180] Urinary CAI excretion in newborns was tracked over the first ten postnatal days to capture time -dependent changes related to development. Based on sample availability, participants were grouped by CAI status measured in days 1-7 relative to days 8-10, yielded three groups: (i) physiological haemolysis with no more than low signal in the first week and CAI -negative urine thereafter; (ii) transient haemolysis with significant signal in the first week and reduced but detectable signal thereafter, and (iii) sustained haemolysis with strong or increasing signal. Signals in the last group reached levels comparable to malaria-positive adult urine samples. Two participants produced too few samples to assign to these groups. Noting that all bar three urine samples were Hb-negative, CAI -positive urines likely reflect a haemolytic event. Grouping, clinical diagnosis, bilirubin, peak CRP during first 10 days, DAT status, and initial and average white blood counts (WBC) and Hb levels are compiled in Figure 6A. Urinary CAI EUISA time courses by grouping (Figures 6B) are shown alongside bilirubin, pooling serum and transcutaneous measurements (Figures 6C). Despite dramatically different urinary CAI excretion profiles in the three groups, bilirubin time courses were not significantly different, even after normalizing to NICE thresholds. Whilst raised bilirubin generally relates to haemolysis, it is neither a specific nor a quantitative marker of a haemolytic crisis, thus the lack of concordance is not unexpected. To determine which metric contributes most to between- participant variation, principal component analysis was performed on standardized data. Urinary CAI projected onto the first principal component and all but one participant in the sustained haemolysis group clustered on the right side of the scatter (Figure 6D). Physiological CAI measurements associated with full-term births, whereas greater CAI excretion correlated with prematurity (Figure 6E). Peak CRP showed considerable variation, which may reflect differences in illness and hepatic maturity (Figure 6F). Nonetheless, there was a significant association between CRP status (threshold 5 mg / L) and haemolysis group (Figure 6F). Physiologically haemolyzing participants tended to have white blood cell (WBC) counts in the range 15-25 billion / L, whereas transient and sustained haemolysis groups had lower WBCs (Figures 6G). Whilst ~15 billion / L is not considered leukopenic, the inverse association between CAI excretion and WBC counts may indicate increased likelihood of neonatal sepsis, particularly with raised CRP. Example 7 - Urinary CAI excretion is accurately measured on a lateral flow device

[0181] A urine-borne biomarker that correlates with disease surrogates is well suited for routine point- of-care testing using lateral flow devices (LFDs; Figure 7A). Two designs were produced, differing in concentration of capture antibodies (five-fold higher in design #2), and used to test a panel of 52 representative samples from all study arms (Figures 7B / C). LFD performance was benchmarked against ELISA (Figure 7D) and western blot (Figure 7E). Sensitivity and specificity were excellent, reaching >90% for design #2 against western blot score. For this analysis, the threshold for a positive T / C ratio was the upper limit of control samples that had a western blot score of 0 or sub-threshold ELISA measurement. Critically, there were very few false-positive LFD results, and a small number of false-negatives, mostly among the Peruvian samples which may reflect epitope attrition due to storage and transport.

[0182] The study presents the case for urinary CAI excretion as a biomarker of haemolysis inside blood vessels, introduces an LFD for point-of-care testing, and describes associations and thresholds that can help interpret CAI signals. A key advantage of urinary CAI excretion is that its appearance is not contingent on developmentally sensitive processing and can be intuitively related to the magnitude of the haemolytic event above a negligible baseline. Testing urine for CAI in the first and second postnatal weeks can classify newborns into haemolysis groups and assess the risk of associated complications, such as sepsis. Screening at-risk adult groups, such as sickle-cell disease patients and people in malaria-endemic regions, can assess haemolysis and test for adverse events, guiding better dosages and choice of drugs, especially under the economic constraint of LMICs and drive for mass drug administration to eradicate malaria. Proactive haemolysis testing could help flag complications following transfusions or bone marrow transplants, screen potentially fatal events such as TTP / HUS, or identify adverse reactions to foods in favism.

Claims

CLAIMS1. A method for determining the presence or absence of haemolysis inside blood vessels in a mammalian subject, comprising a step of detecting carbonic anhydrase 1 (CAI) in a urine sample from the subject, wherein the subject is not a human neonate.

2. The method of claim 1, wherein the subject is a human, a dog, a cat or a horse.

3. The method of claim 1 or claim 2, wherein the subject is a human who is at least 29 days old, at least 1 year old, at least 5 years old, at least 10 years old, at least 15 years old or at least 18 years old.

4. A method for determining the presence or absence of haemolysis inside blood vessels in a human neonate who does not meet the threshold for phototherapy according to National Institute for Health and Care Excellence (NICE) guidelines, comprising a step of detecting carbonic anhydrase 1 (CAI) in a urine sample from the human neonate.

5. The method of any preceding claim, wherein: a. CAI is quantified; and / or b. the method further comprises a step of detecting haemoglobin (Hb); and / or c. the method comprises quantifying Hb; and / or d. the method further comprises a step of comparing the concentration of the one or more marker(s) to a reference.

6. The method of any preceding claim wherein the marker is quantified using an ELISA assay and / or a western blot assay and / or a lateral flow immunoassay.

7. The method of any preceding claim, wherein the subject is a human suffering from a disease selected from the group consisting of Bartonella causing Carrion’s disease, Clostridium perfringens, enterohaemorrhagic strains of Escherichia coli, Shigella dysenteriae, Mycoplasma pneumoniae, Rickettsia rickeiisii. Salmonella enterica, Streptococcus pneumoniae, West Nile virus, Parvovirus B19, Ebstein-Barr virus, human immunodeficiency virus, morbillivirus, paramyxoviruses, rubivirus, varicella-zoster virus, cytomegalovirus, hepatitis A virus, dengue virus, Flaviviridae, Filoviridae, Arenaviridae or Bunyaviridae, Plasmodium causing malaria, Trichomonas vaginalis causing trichomoniasis, Leishmania causing leishmanioses, Trypanosoma causing trypanosomal disease, Babesiacausing babesiosis, Entamoeba histolytica, sickle cell disease, thalassemia, hereditary spherocytosis, hereditary elliptocytosis, glucose 6-phosphate dehydrogenase deficiency (G6PDD), pyruvate kinase deficiency (PKD), glutathione synthetase deficiency (GSD), rhesus disease, HELLP syndrome, thrombotic thrombocytopenic purpura (TTP) and haemolytic uremic syndrome (HUS).

8. The method of any one of claims 1 to 6, wherein the subject suffers from exertion -induced haemolysis.

9. A method of determining the presence or absence of an infective, immunological, pharmacological, biochemical, or food or beverage trigger of haemolysis in a subject, comprising a step of detecting carbonic anhydrase 1 (CAI) in a urine sample from the subject.

10. A method for determining severity of haemolysis in a mammalian subject, comprising a step of detecting carbonic anhydrase 1 (CAI) in a urine sample from the subject.

11. A method for determining the efficacy, safety and / or dose -escalation of a drug treatment in a mammalian subject, comprising a step of detecting carbonic anhydrase 1 (CAI) in a urine sample from the subject.

12. The method of claim 11, wherein the patient suffers from a disease selected from the group consisting Bartonella causing Carrion’s disease, Clostridium perfringens, enterohaemorrhagic strains of Escherichia coli, Shigella dysenteriae, Mycoplasma pneumoniae, Rickettsia rickettsii, Salmonella enterica, Streptococcus pneumoniae, West Nile virus, Parvovirus Bl 9, Ebstein-Barr virus, human immunodeficiency virus, morbillivirus, paramyxoviruses, rubivirus, varicella-zoster virus, cytomegalovirus, hepatitis A virus, dengue virus, Flaviviridae, Filoviridae, Arenaviridae or Bunyaviridae, Plasmodium causing malaria, Trichomonas vaginalis causing trichomoniasis, Leishmania causing leishmanioses, Trypanosoma causing trypanosomal disease, Babesia causing babesiosis, Entamoeba histolytica, sickle cell disease, thalassemia, hereditary spherocytosis, hereditary elliptocytosis, glucose 6-phosphate dehydrogenase deficiency (G6PDD), pyruvate kinase deficiency (PKD), glutathione synthetase deficiency (GSD), rhesus disease, HELLP syndrome, thrombotic thrombocytopenic purpura (TTP) and haemolytic uremic syndrome (HUS).

13. The method of claim 12, wherein the patient suffers from malaria.3614. The method of any one of claims 11-13 wherein the method comprises quantifying CAI; and / or the method further comprises a step of comparing the concentration of CAI to a reference value.

15. A method of distinguishing between haemolysis inside blood vessels and blood contamination of a sample, comprising a step of detecting CAI and Hb in the sample.

16. The method of claim 15, wherein the presence of CAI and absence of Hb indicates haemolysis inside blood vessels or the presence of CAI and presence of Hb indicates blood contamination.

17. The method of any of the preceding claims wherein CAI is quantified using an ELISA assay and / or a western blot assay and / or a lateral flow immunoassay.

18. The method of claim 17 wherein a concentration above the reference value indicates the presence of haemolysis, or a concentration equal to or below the reference value indicates the absence of haemolysis.

19. The method of any preceding claim, wherein the method is performed ex vivo or in vitro.

20. The method of any preceding claim, wherein the haemolysis inside blood vessels is intravascular haemolysis.

21. A kit comprising two or more antibodies wherein at least one antibody can detect CAI and at least one antibody can detect Hb.

22. The kit of claim 21, wherein the two or more antibodies are monoclonal, polyclonal or recombinant antibodies, and / or wherein the kit comprises a lateral flow immunoassay.

23. A lateral flow assay device comprising: (i) a sample receiving region; and (ii) a capture membrane positioned downstream of the sample receiving region and comprising a test region including an antibody suitable for detecting CAI.

24. The lateral assay device of claim 23, wherein the capture membrane further comprises an antibody suitable for detecting Hb.

25. Use of the kit of any one of claims 21 to 22 or the lateral flow device of any one of claims23 to 24 in a method of determining the presence or absence of haemolysis in a subject.

26. The kit for use according to claim 25, wherein the haemolysis is haemolysis inside blood vessels or bleeding along the urogenital tract or menstruation.

Images