Supplementary Material:
Clinical utility and implications of asparaginase antibodies in acute lymphoblastic leukemia
Chengcheng Liu,1,5 Jitesh D. Kawedia, PhD1 Cheng Cheng, PhD2 Deqing Pei, 2 Christian A. Fernandez, PhD1 Xiangjun Cai, PhD1 Kristine R. Crews, PharmD1 Sue C. Kaste, DO3,5 John C. Panetta, PhD1 W. Paul Bowman, MD6 Sima Jeha, MD4 John T. Sandlund, MD4 William E. Evans, PharmD1 Ching-Hon Pui, MD4 and Mary V. Relling, PharmD1
From Departments of 1Pharmaceutical Sciences, 2Biostatistics, 3Radiological Sciences, and 4Oncology, St Jude Children’s Research Hospital, Memphis, TN; 5University of Tennessee Health Science Center, Memphis, TN; and 6Cook Children’s Hospital, Fort Worth, TX.
Correspondence: Mary V. Relling, St. Jude Children’s ResearchHospital, 262 Danny Thomas Place, Memphis, TN 38105-2794, USA. Phone: 901-595-2348; Fax: 901-595-8869; e-mail:
Supplement
- Supplemental Methods
- Supplemental Tables
Table S1. Clinical features of patients with and without samples evaluable for anti-asparaginase antibodies.
Table S2.Number of available blood samples at each time point.
Table S3.Number of patients having allergic reaction to Elspar in different phases.
Table S4. Clinical features and allergic reactions to Elspar.
Table S5. Clinical features and anti-Elspar antibody status.
- Supplemental Figures
Figure S1.Total XV asparaginase regimen and anti-asparaginase antibody measurements.
Figure S2. Anti-Elspar antibody level relative to the time of clinical reaction to Elspar.
Figure S3.ROC curves of the antibody tests.
Figure S4. Association between Week 7 anti-Elspar antibody level and the proportion of patients reacted to Elspar around week 7.
Figure S5. The frequency of hypersensitivity to Oncaspar and Erwinase.
- Supplemental References
Supplemental methods
Anti-asparaginase antibodies
A total of 2010 serum samples for antibodies to three forms of asparaginase (Elspar, Erwinase, and Oncaspar) were measured by ELISA. Of these, 471 samples were assayed at the University of Tennessee Health Science Center (UTHSC) and 1345 samples were assayed at St. Jude Children’s Research Hospital (SJCRH). The positive control reference pool was prepared from patients known to have received the specific form of asparaginase and who exhibited clinical signs of allergic reaction and had high titers of antibodies, represented by high optical density (OD) readings. The negative control reference pool was prepared from healthy volunteers who had never received asparaginase treatment. A sample assayed at SJCRH was considered positive if the natural log of its 1:400 OD reading was greater than 2.58 standard deviations above the negative control process mean. A sample assayed at UTHSC was considered positive if the natural log of its 1:3200 OD reading was greater than 2 standard deviations above the negative control process mean. Samples that were not positive were considered negative. Because the magnitudes of the scale of antibody OD readings obtained at the two facilities differed, we normalized the OD readings based on a comparison of 194 serum samples measured for antibody levels at both SJCRH and UTHSC. The normalization of UTHSC OD readings to the same scale was applicable to the SJCRH readings was described by an exponential model using equation 1 (Elspar), 2 (Erwinase), and 3 (Oncaspar) to maintain consistency of OD readings between UTHSC and SJCRH. All of the samples measured at UTHSC had their OD readings normalized to the scale of the readings obtained at SJCRH using the following equations.
log (normalized OD) = -5.066 + 3.509 ∙ exp [0.526 × log (original OD)][1]
log (normalized OD) = -4.360 + 3.518 ∙ exp [0.806 × log (original OD)] [2]
log (normalized OD) = -5.280 + 3.684 ∙ exp [0.520 × log (original OD)] [3]
Estimation of antibody area-under-the-curve (AUC)
The area under the antibody-concentration-versus-time curve (AUC) was estimated using the method of trapezoids in the 360 patients who had at least four out of five scheduled antibody results (including week 7 and week 17 tests in the continuation phase and at least two antibody tests in the induction phase). Induction AUC (antibody AUC between day 5 and day 34) and continuation AUC (antibody AUC between week 7 and week 17) were estimated separately using OD readings for each patient’s serum tested against Elspar, the front-line asparaginase preparation (Supplemental Figure 1). Analyses were performed using the sum of the induction AUC and continuation AUC of antibodies to Elspar asparaginase.
Asparaginase activity
This assay was similar to a previously reported assay,1 with a faster and more controlled process as well as improved linearity range without sample pretreatments. The asparaginase activity in serum was measured by an enzymatically coupled oxidation of reduced nicotinamide adenine dinucleotide (NADH):
L-Asparagine L-Aspartate + NH4
L-Aspartate + α-Ketoglutarate Oxaloacetate + L-Glutamate
Oxaloacetate + NADH Malate + NAD+
When monitored by spectrophotometry, the oxidation of NADH leads to a corresponding decrease in absorbance at 340 nM. The rate of change of absorbance is directly proportional to asparaginase activity.
The enzyme reaction mixture was prepared by combining 200 mL of 10x Tris-buffered saline, 400 mL of glycerol, 200 mg of α-ketoglutaric acid, 264 mg of asparagine, 200 mg of β-NADH, 400 µL of glutamic oxaloacetic transaminase,200 µL of malic dehydrogenase,and 1400 mL distilled and deionized water, and stored at -80ºC. (All the reagents above were from Sigma-Aldrich, St. Louis, MO.) Before plate reading, the enzyme reaction mixture was incubated in a 37ºC water bath for 30 min. Then patient sera were added to 96-well UV-transparent plates (Corning. Inc, NY) in quadruplicate. The reaction was started by adding 190 µL of enzyme reaction mixture to each well. The enzyme reaction mixture was added one row at a time, with a 20 second interval between each addition. The plate was read using a BioTek ELx808IU Microplate Reader (BioTek, Winooski, VT) at 340 nm with a 20 second interval for 15 min. The activity was estimated by the slope of the change of optical density over time. The linearity range of this assay was 0.011-2.2 U/mL. Samples higher than 2.2 U/mL were diluted with sample dilution buffer (5% bovine serum albumin in deionized water).Calibrators were established with asparaginase standard solutions: 0.025-2.0 U/mL native E. coli asparaginase(BioVendor Laboratory Medicine.Inc, Czech Republic) in sample dilution buffer.
Receiver-operator-characteristic (ROC) curves
The ROCcurves of antibody tests were generated by plotting the true positive rate (sensitivity) and corresponding false positive rate (1-specificity) for every possible threshold OD value (to separate positive from negative samples; supplemental Figure 3). The area-under-the-curve (AUC) of the ROC curve is a relative measure of the diagnostic accuracy of a test. An AUC of 1.0 would indicate a perfect test (100% TP, 0% FP, 100% TN, 0% FN), whereas 0.5 would indicate a random classification. A value greater 0.9 represents a rather high accuracy, a value less than 0.7 represents a low accuracy, and a value in between would be considered fair to good.2 (TP, true positive; FP, false positive; TN, true negative; FN, false negative.)
Supplemental Tables.
Table S1. Clinical features of patients with (n = 410) and without (n = 88) samples evaluable for anti-asparaginase antibodies.
Patients with samples n (%) / Patients without samples n (%) / PAge
Less than 10 years / 302 (81.2) / 70 (18.8) / 0.28
Older than 10 years / 108 (85.7) / 18 (14.3)
Sex
Male / 228 (81.7) / 51 (18.3) / 0.72
Female / 182 (83.1) / 37 (16.9)
Racea
White / 282 (88.4) / 37 (11.6) / 0.0001
Black / 65 (91.5) / 6 (8.5)
Other / 63 (58.3) / 45 (41.7)
Immunophenotypeb
B-lineage / 345 (81.6) / 78 (18.4) / 0.33
T-lineage / 65 (86.7) / 10 (13.3)
Treatment arm
Low-risk / 197 (82.4) / 42 (17.6) / 1.0
Standard/high-risk / 213 (82.2) / 46 (17.8)
a = Genetically determined race as described;3 b = acute lymphoblastic leukemia immunophenotype.
Table S2. Number of patients with evaluable blood samples at each time point.
Day 5 / 401 (420)
Day 19 / 392 (396)
Day 34 / 384 (399)
Week 7 / 381 (408)
Week 17 / 367 (387)
Table S3.Number of patients having allergic reaction to Elsparat each phase of therapy.
Phase / Low-risk / Standard/high-risk / TotalInduction / 2 / 3 / 5
Continuation weeks 1-6 / 0 / 54 / 54
Reinduction I / 96 / 1 / 97
Continuation weeks 10-16 / 0 / 7 / 7
Reinduction II / 2 / 0 / 2
Reintensification / 0 / 4 / 4
Total / 100 / 69 / 169
Table S4. Clinical features and allergic reactions to Elspar (n=410*)
All patients / Low-risk Arm / Standard/high-risk Armn / Allergy n(%) / No allergy n(%) / P / n / Allergy n(%) / No allergy n(%) / P / n / Allergy n(%) / No allergy n(%) / P
Age
Less than 10 years / 302 / 129 (42.7) / 173 (57.3) / 0.36 / 183 / 94(51.4) / 89(48.6) / 0.59 / 119 / 35(29.4) / 84(70.6) / 0.31
Older than 10 years / 108 / 40 (37.0) / 68 (63.0) / 14 / 6 (42.9) / 8 (57.1) / 94 / 34 (36.2) / 60 (63.8)
Sex
Male / 228 / 96 (42.1) / 132 (57.9) / 0.69 / 93 / 53(57.0) / 40(43.0) / 0.12 / 135 / 43(31.9) / 92(68.1) / 0.88
Female / 182 / 73 (40.1) / 109 (59.9) / 104 / 47 (45.2) / 57 (54.8) / 78 / 26 (33.3) / 52 (66.7)
Racea
White / 282 / 128 (45.4) / 154 (54.6) / 0.037 / 146 / 75 (51.4) / 71 (48.6) / 0.19 / 136 / 53 (39.0) / 83 (61.0) / 0.004
Black / 65 / 20 (30.8) / 45 (69.2) / 24 / 15 (62.5) / 9 (37.5) / 41 / 5 (12.2) / 36 (87.8)
Other / 63 / 21 (33.3) / 42 (66.7) / 27 / 10 (37.0) / 17 (63.0) / 36 / 11 (30.6) / 25 (69.4)
Immunophenotypeb
B-lineage / 345 / 157 (45.5) / 188 (54.5) / .0001 / 197 / 100 (50.8) / 97 (49.2) / NA / 148 / 57(39.2) / 91(60.8) / 0.003
T-lineage / 65 / 12 (18.5) / 53 (81.5) / 0 / 0 (0.0) / 0 (0.0) / 65 / 12 (18.5) / 53 (81.5)
a = Genetically determined race as described;3 b = acute lymphoblastic leukemia immunophenotype; NA = not applicable.
*Patients with serum samples evaluable for anti-asparaginase antibodies (n = 410). Analysis for all patients(n=498) has been previously reported.4
Table S5. Clinical features and anti-Elspar antibody status (n=410)
All patients / Low-risk / Standard/high-riskn / Positive n(%) / Negative n(%) / P / n / Positive n(%) / Negative n(%) / P / n / Positive n(%) / Negative n(%) / P
Age
Less than 10 years / 302 / 182 (60.3) / 120 (39.7) / 0.07 / 183 / 129 (70.5) / 54 (29.5) / 0.14 / 119 / 53 (44.5) / 66 (55.5) / 0.49
Older than 10 years / 108 / 54 (50.0) / 54 (50.0) / 14 / 7 (50.0) / 7 (50.0) / 94 / 47 (50.0) / 47 (50.0)
Sex
Male / 228 / 135 (59.2) / 93 (40.8) / 0.48 / 93 / 70 (75.3) / 23 (24.7) / 0.09 / 135 / 65 (48.2) / 70 (51.8) / 0.67
Female / 182 / 101 (55.5) / 81 (44.5) / 104 / 66 (63.5) / 38 (36.5) / 78 / 35 (44.9) / 43 (55.1)
Racea
White / 282 / 175 (62.1) / 107 (37.9) / 0.005 / 146 / 104 (71.2) / 42 (28.8) / 0.40 / 136 / 71 (52.2) / 65 (47.8) / 0.013
Black / 65 / 36 (55.4) / 29 (44.6) / 24 / 16 (66.7) / 8 (33.3) / 41 / 20 (48.8) / 21 (51.2)
Other / 63 / 25 (39.7) / 38 (60.3) / 27 / 16 (59.3) / 11 (40.7) / 36 / 9 (25.0) / 27 (75.0)
Immunophenotypeb
B-lineage / 345 / 213 (61.7) / 132 (38.3) / .0001 / 197 / 136 (69.0) / 61 (31.0) / NA / 148 / 77 (52.0) / 71 (48.0) / 0.026
T-lineage / 65 / 23 (35.4) / 42 (64.6) / 0 / 0 (0.0) / 0 (0.0) / 65 / 23 (35.4) / 42 (64.6)
a = Genetically determined race as described;3 b = acute lymphoblastic leukemia immunophenotype; NA = not applicable.
Supplementalfigure legends
Figure S1. Total XV asparaginase regimen and anti-asparaginase antibody measurements. All patients received Elspar (ASP) at 10000 U/m2 per dose during remission induction; patients on the SHR arm received 25000 U/m2 per week at weeks 1-19; those on the LR arm received 10000 U/m2 thrice weekly at reinduction weeks 7-9 and 17-19. Serum samples for anti-asparaginase antibodies (ANTI-ASP) were collected on days 5, 19, and 34 of remission induction and weeks 7 and 17 of continuation therapy.The area under the antibody-concentration-versus-time curve (ANTI-ASP AUC) was estimated forinduction (between day 5 and day 34) and continuation (between week 7 and week 17) using OD readings tested against Elspar.
Figure S2. Anti-Elspar antibody level relative to the time of clinical reaction to Elspar. Anti-Elspar OD level in patient samples (dots) relative to the number of days elapsed before (negative) or after (positive) between sample collection and clinical reaction to Elspar. The trend lines of median (solid) and quartiles (dashed) are shown.
Figure S3. ROC curves of the antibody tests. Receiver operating characteristic (ROC) curves of the antibody tests at day 34, week 7 and week 17, showing the true positive rates and respective false positive rates for predicting or confirming clinical allergic reactions. Each point on the curve corresponds to a specific threshold of anti-Elspar OD that putatively divides antibody-positive from antibody-negative samples. Right y-axis shows the color key that visualizes the mapping from the natural log of anti-Elspar OD to colors. The positive threshold dividing positive from negative samples for logOD is -3.2 (circles). The AUC of the ROC curves, which represents overall test performance on predictions, is shown in each graph. LR, low-risk; SHR, standard/high-risk; Ab, anti-Elspar antibody; Rxn, clinical reaction; AUC, area under the curve. Time of antibody tests: D34 = Day 34 of induction therapy; W7 = Week 7 of continuation therapy; and W17 = Week 17 of continuation therapy.
Figure S4. Association between Week 7 anti-Elspar antibody level and the proportion of patients reacting to Elspar around week 7. Left panels: the distribution of patients with (solid bars) and without (open bars) clinical reactions by their anti-Elspar OD at week 7. (a) Patients on the LR arm (n=195) with and without reactions during weeks 7-9. (c) Patients on the SHR arm (n=175) with and without reactions during weeks 1-6. Right panels: the positive correlation between week 7 anti-Elspar OD and the proportion of patients having clinical reaction at (b) weeks 7-9 (LR) and (d) weeks 1-6 (SHR). Trend lines are shown. LR, low-risk; SHR, standard/high-risk.
Figure S5.The frequency of hypersensitivity to Oncaspar and Erwinase. The proportion of patients exhibiting secondary allergic reactions and antibody positivityamong the patients who had samples tested for anti-Oncaspar (n=85) or anti-Erwinase (n=63) after they received Oncaspar or Erwinase. Ab, antibody against Oncaspar (left) or Erwinase (right).
References
1.Asselin BL, Whitin JC, Coppola DJ, Rupp IP, Sallan SE, Cohen HJ. Comparative pharmacokinetic studies of three asparaginase preparations. J Clin Oncol 1993; 11: 1780-1786.
2.Swets JA. Measuring the accuracy of diagnostic systems. Science 1988; 240: 1285-1293.
3.Yang JJ, Cheng C, Devidas M, Cao X, Fan Y, Campana D, et al. Ancestry and pharmacogenomics of relapse in acute lymphoblastic leukemia. Nat Genet 2011; 43: 237-241.
4.Chen SH, Pei D, Yang W, Cheng C, Jeha S, Cox NJ, et al. Genetic variations in GRIA1 on chromosome 5q33 related to asparaginase hypersensitivity. Clin Pharmacol Ther 2010; 88: 191-196.
1