Method for predicting the outcome of a treatment with aflibercept of a patient suspected to suffer from a cancer

Abstract

The present invention concerns the use of interleukin-8 (IL-8) as a biomarker for predicting the outcome of the treatment with aflibercept, or ziv-aflibercept of a patient suspected to suffer from a cancer.

Claims

1. A method of determining the therapeutic efficacy of aflibercept or ziv-aflibercept in a patient with colon cancer, colorectal cancer, or rectal cancer, comprising subjecting a biological sample from the patient to at least one assay to measure the IL-8 expression level in the biological sample, and comparing the measured IL-8 expression level to a reference level of IL-8, wherein an IL-8 expression level measured in the biological sample lower than the reference level is indicative of the therapeutic efficacy of aflibercept in said patient.

2. The method of claim 1, wherein the colon cancer is metastatic colon cancer.

3. The method of claim 1, wherein the colorectal cancer is metastatic colorectal cancer.

4. The method of claim 1, wherein the rectal cancer is metastatic rectal cancer.

5. The method of claim 1, wherein the reference level of IL-8 is between about 10 and about 30 pg/mL.

6. The method of claim 1, wherein the reference level of IL-8 is about 19 pg/mL.

7. The method of claim 1, wherein the biological sample is selected from the group consisting of blood, serum and plasma.

8. The method of claim 1, wherein the IL-8 level is a circulating level.

9. The method of claim 1, wherein the patient has previously been treated with oxaliplatin or bevacizumab.

10. The method of claim 1, wherein the reference level of IL-8 is the expression level of IL-8 in a subject without cancer.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

(1) FIGS. 1 and 2 illustrate the relation between IL8 levels and the probability of disease progression. Depicted is the probability of disease progression after 12 months in relation to IL-8 plasma levels at baseline (FIG. 1) and the difference between IL8 plasma levels at baseline and the last measurement point before disease progression (FIG. 2). Briefly, FIG. 1 shows that high IL8 levels correlate with increased probability for disease progression and that this effect is slightly more pronounced in aflibercept versus FOLFOX treated patients. On the other hand, FIG. 2 shows that the increase in IL8 relative to baseline also corresponds to disease progression. Since increases in IL8 are plotted on a logarithmic scale, values between −3 and 0 correspond to an increase in IL8<1 pg/mL, whereas values between 0 and 3 represent increases>1 pg/mL. Data thus show that even small increases in IL8 relative to baseline already correspond to an increased probability of disease progression in the aflibercept arm.

DETAILED DESCRIPTION OF THE INVENTION

Example: Effect of Interleukin 8 on PFS in the AFFIRM Study

Study ECF10668 (AFFIRM)

(2) EFC10668 was designed as a randomized, multinational, study comparing the adverse effects occurrence in patient with metastatic colorectal cancer (MCRC) treated with:

(3) i) a modified FOLFOX6 (a combination of oxaliplatin, 5-fluorouracil (5-FU) and folinic acid) given intravenously every 2 weeks as first-line treatment (arm A); or

(4) ii) aflibercept at 4 mg/kg combined with a modified FOLFOX6 given intravenously every 2 weeks as first-line treatment; or

Schedule of Administration

(5) Patients were administered intravenously either with aflibercept immediately followed by oxaliplatin, 5-fluorouracil (5-FU) and folinic acid (modified FOLFOX6 regimen) or modified FOLFOX6 alone, depending on arm to which they were assigned,

(6) This treatment was repeated every 2 weeks until progression (or unacceptable toxicity, or consent withdrawal).

Dosage

(7) The patients randomized in the aflibercept arm received 4 mg/kg IV every 2 weeks.

(8) The following were administered to patients in both treatment groups: Oxaliplatin (Eloxatin®) Folinic acid (also known as leucovorin) 5-fluorouracil

(9) Formulations of oxaliplatin, 5-fluorouracil, and folinic acid: Products used were those available in the hospital/clinic pharmacy Route of administration: IV

(10) Dose: Oxaliplatin, folinic acid, and 5-fluorouracil were administered according to an mFOLFOX6 regimen, as follows: Oxaliplatin 85 mg/m.sup.2 as a 2-hour IV infusion on day 1 Folinic acid 350 mg/m.sup.2 as a 2-hour IV infusion on day 1 5-fluorouracil 400 mg/m.sup.2 as an IV bolus on day 1, and then 2400 mg/m.sup.2 as a 46-hour continuous IV infusion starting on day 1

(11) In case of body surface area>2.0 m.sup.2, the actual doses of oxaliplatin and of 5-FU were to be adjusted to a maximum BSA of 2.0 m.sup.2 for safety reasons. Dose reduction and/or treatment delay and/or treatment discontinuation were planned in case of severe toxicity. The modified FOLFOX6 regimen was administered after administration of aflibercept.

Duration of Treatment

(12) Treatment for an individual patient was administered up until progression or until unacceptable toxicity occurred or the patient withdrew consent.

(13) Treatment duration was estimated to be approximately 12 months.

Demographics and Baseline Characteristics

(14) Table 1 below compares demographics and patient characteristics at baseline between biomarkers evaluable and non-evaluable populations.

(15) The “biomarkers evaluable population” is defined as the population of patients who provided a blood/tumor sample for biomarker assessment; while the “biomarkers non evaluable population” corresponds to patients who did not provide blood/tumor sample for biomarker assessment (e.g. patients who did not consent to biomarker study).

(16) All characteristics are similar between populations, except for the region of origin of the patients: Eastern Europe tends to be over-represented and other countries tend to be under-represented in the biomarkers evaluable population compared to the biomarker non evaluable population.

(17) TABLE-US-00001 TABLE 1 Summary of patient demographics and patient characteristics at baseline-Evaluable population Biomarkers non Biomarkers evaluable evaluable population population Aflibercept/ Aflibercept/ mFolfox6 mFolfox6 mFolfox6 mFolfox6 (N = 57) (N = 49) (N = 60) (N = 70) p-value Gender 1.0000.sup.a [n(%)] Number 57 49 60 70 Male 32 33 36 43 (56.1%) (67.3%) (60.0%) (61.4%) Female 25 16 24 27 (43.9%) (32.7%) (40.0%) (38.6%) Age (Years) 0.2811.sup.b Number 57 49 60 70 Median 66.0 62.0 62.0 62.5 Mean 63.7 61.8 61.3 61.7 (SD) (10.0) (9.5) (9.4) (8.7) Min:Max 44:87 29:75 37:81 41:79 Age class 0.2421.sup.a [n(%)] Number 57 49 60 70 <65 27 28 38 42 (47.4%) (57.1%) (63.3%) (60.0%) ≥65 but 23 20 20 25 <75 (40.4%) (40.8%) (33.3%) (35.7%) ≥75 7 (12.3%) 1 (2.0%) 2 (3.3%) 3 (4.3%) Race [n(%)] 0.0640.sup.a Number 57 49 60 70 Caucasian/ 41 37 49 60 White (71.9%) (75.5%) (81.7%) (85.7%) Black 0 0 0 1 (1.4%) Asian/ 16 11 11 9 Oriental (28.1%) (22.4%) (18.3%) (12.9%) Other 0 1 (2.0%) 0 0 Region 0.0029.sup.a Number 57 49 60 70 Western 37 30 37 45 Europe (64.9%) (61.2%) (61.7%) (64.3%) Eastern 3 (5.3%) 3 (6.1%) 12 12 Europe (20.0%) (17.1%) Other 17 16 11 13 countries (29.8%) (32.7%) (18.3%) (18.6%) BSA (m2) 0.1091.sup.b Number 57 49 60 70 Median 1.8 1.8 1.8 1.8 Mean 1.7 (0.2) 1.8 (0.2) 1.8 (0.2) 1.8 (0.2) (SD) Min:Max 1:2 1:2 1:2 1:2 Weight (kg) 0.0838.sup.b Number 57 49 60 70 Median 67.6 70.0 73.2 71.4 Mean 67.3 71.1 74.0 71.5 (SD) (14.1) (16.6) (17.0) (15.6) Min:Max 40:107 40:115 48:134 40:117 .sup.acomparing frequency distribution based on Fisher's exact test-2-sided. .sup.bUsing ANOVA (type 3) with factors: BIOPOP, BIOPOP. Records with missing values for factors or response were excluded from statistical analyses. Frequency distribution of covariates is compared between evaluable and non-evaluable populations Note: Western Europe = Germany, Italy, Spain, United Kingdom; Eastern Europe = Russian Federation; Other countries = Australia, Korea

Disease Characteristics at Baseline

(18) Disease characteristics at baseline were similar in the two populations (see Tables 2 and 3 below).

(19) TABLE-US-00002 TABLE 2 Summary of disease characteristics at initial diagnosis Biomarkers Biomarkers non evaluable evaluable population population Aflibercept/ Aflibercept/ mFolfox6 mFolfox6 mFolfox6 mFolfox6 (N = 57) (N = 49) (N = 60) (N = 70) p-value Primary site 0.7045.sup.a [n(%)] Number 57 49 60 70 Colon 27 26 (53.1%) 31 33 (47.1%) (47.4%) (51.7%) Recto 9 15 (30.6%) 16 19 (27.1%) sigmoid (15.8%) (26.7%) Rectum 21 8 (16.3%) 13 18 (25.7%) (36.8%) (21.7%) Histology type [n(%)] Number 57 49 60 70 Adeno- 57 49 60 70 (100%) carcinoma (100%) (100%) (100%) Staging at 0.2297.sup.a diagnosis [n(%)] Number 56 48 57 68 Stage I 1 (1.8%) 0 2 (3.5%) 2 (2.9%) Stage II 2 (3.6%) 1 (2.1%) 6 4 (5.9%) (10.5%) Stage III 5 (8.9%) 3 (6.3%) 3 (5.3%) 4 (5.9%) Stage IV 48 44 46 58 (85.7%) (91.7%) (80.7%) (85.3%) Time from 0.6620.sup.b diagnosis to random- ization (months)* Number 57 49 60 69 Median 1.4 1.8 1.6 1.7 Mean (SD) 12.6 9.2 (16.9) 9.8 9.7 (17.5) (30.9) (19.8) Min:Max 0:149 0:80 0:84 0:80 .sup.acomparing frequency distribution based on Fisher's exact test-2-sided. .sup.bUsing ANOVA (type 3) with factors: BIOPOP, BIOPOP. Records with missing values for factors or response were excluded from statistical analyses. Frequency distribution of covariates is compared between evaluable and non-evaluable populations *If the day of initial date of diagnosis is missing, it is considered as the first day of the month

(20) TABLE-US-00003 TABLE 3 Summary of organs involved at baseline Biomarkers Biomarkers non evaluable evaluable population population Aflibercept/ Aflibercept/ mFolfox6 mFolfox6 mFolfox6 mFolfox6 (N = 57) (N = 49) (N = 60) (N = 70) p-value Number of 0.1711.sup.a metastatic organs involved at baseline (excluding primary site) [n(%)] Number 57 49 60 70 0 0 0 1 (1.7%) 0 1 16 (28.1%) 15 (30.6%) 15 12 (17.1%) (25.0%) >1 41(71.9%) 34 (69.4%) 44 58 (82.9%) (73.3%) Metastatic 57 (100%) 49 (100%) 59 70 (100%) organs (98.3%) involved at baseline (excluding primary site) [n(%)]* Liver 44 (77.2%) 42 (85.7%) 47 57 (81.4%) (78.3%) Lung 27 (47.4%) 18 (36.7%) 25 28 (40.0%) (41.7%) Lymph nodes 26 (45.6%) 25 (51.0%) 30 38 (54.3%) (50.0%) Muscle/ 9 (15.8%) 6 (12.2%) 14 10 (14.3%) soft tissue (23.3%) Peritoneum 8 (14.0%) 7 (14.3%) 8 16 (22.9%) (13.3%) Pleura 5 (8.8%) 7 (14.3%) 2 (3.3%) 1 (1.4%) Adrenal 2 (3.5%) 0 1 (1.7%) 1 (1.4%) Bone 2 (3.5%) 1 (2.0%) 4 (6.7%) 4 (5.7%) Kidneys 1 (1.8%) 0 0 0 Spleen 1 (1.8%) 1 (2.0%) 1 (1.7%) 1 (1.4%) Bladder 0 0 1 (1.7%) 1 (1.4%) Metastatic 0.3536.sup.a organs involved at baseline class (excluding primary site) [n(%)] Number 57 49 60 70 No liver 49 (86.0%) 39 (79.6%) 51 63 (90.0%) metastasis, (85.0%) or liver and other metastases Liver 8 (14.0%) 10 (20.4%) 9 7 (10.0%) metastasis (15.0%) only .sup.acomparing frequency distribution based on Fisher's exact test-2-sided. Records with missing values for factors or response were excluded from statistical analyses. Frequency distribution of covariates is compared between evaluable and non-evaluable populations *Percentages are not additive (sum greater than 100%)

Safety Evaluation

A. Extent of Exposure

(21) Table 4 below shows that patients in the biomarkers evaluable population were exposed slightly longer to treatment than patients in the biomarkers non-evaluable population (median number of cycles: 12 versus 9 or 10).

(22) There was no difference in exposure between treatment arms in the biomarkers evaluable population.

(23) TABLE-US-00004 TABLE 4 Summary of overall study treatment exposure Biomarkers non Biomarkers evaluable evaluable population population Aflibercept/ Aflibercept/ mFolfox6 mFolfox6 mFolfox6 mFolfox6 (N = 56) (N = 49) (N = 60) (N = 70) Number of cycles received by patient Sum 614 572 770.0 865.0 Mean (SD) 11.0 (7.0) 11.7 (9.5) 12.8 (7.0) 12.4 (7.9) Median 10.0 9.0 12.0 12.0 Min:Max 1:43 1:44 3:33 1:42 Number of cycles received by patient 1 2 (3.6%) 3 (6.1%) 0 2 (2.9%) 2 2 (3.6%) 2 (4.1%) 0 2 (2.9%) 3 3 (5.4%) 1 (2.0%) 2 (3.3%) 2 (2.9%) 4 3 (5.4%) 4 (8.2%) 5 (8.3%) 8 (11.4%) 5 1 (1.8%) 0 1 (1.7%) 2 (2.9%) 6 1 (1.8%) 6 (12.2%) 0 4 (5.7%) 7 1 (1.8%) 5 (10.2%) 1 (1.7%) 3 (4.3%) 8 6 (10.7%) 1 (2.0%) 11 (18.3%) 2 (2.9%) 9 6 (10.7%) 3 (6.1%) 0 5 (7.1%) 10 5 (8.9%) 2 (4.1%) 4 (6.7%) 3 (4.3%) 11-15 16 (28.6%) 11 (22.4%) 21 (35.0%) 13 (18.6%) 16-20 7 (12.5%) 4 (8.2%) 6 (10.0%) 14 (20.0%) 21-25 1 (1.8%) 3 (6.1%) 5 (8.3%) 6 (8.6%) >25 2 (3.6%) 4 (8.2%) 4 (6.7%) 4 (5.7%) Duration of exposure (weeks) Number 56 49 60 70 Mean (SD) 25.2 (16.0) 27.5 (22.4) 29.5 (16.4) 28.7 (18.7) Median 24.1 23.1 27.3 25.4 Min:Max 2:95 2:106 6:77 2:88 Duration of exposure = ((First date of last cycle + 14)-First date of first cycle)/7 SD: standard deviation

B. Plasma Profiling

(24) The plasma concentration of 27 cytokines, growth factors or soluble receptors was determined by enzyme-linked immunosorbent assays (ELISA) using two Fluorokine® MAP kits (the human angiogenesis panel A and the human high sensitivity cytokine panel; R&D Systems). Competition experiments were conducted to test interference of aflibercept with the detection of VEGF-A, VEGF-D and placental growth factor (PIGF). Angiopoietin-2 (ANGPT2), SDF1-α, HGF, VEGF-C, soluble VEGF receptor 3 (sFLT4, sVEGFR3) and sVEGFR2 were assessed by single ELISA (R&D Systems). Plasma markers were analyzed at baseline, at 30 and 60 days after the first study treatment infusion and 30 days after the last aflibercept infusion.

Statistical Analysis

(25) Differences between patients with evaluable biomarkers and patients without evaluable biomarkers were assessed using a two-sided Fisher's exact test for categorical variables and ANOVA for continuous variables. Biomarkers were analyzed as quantitative variables, by coding the absence or presence of a somatic mutation as 0 or 1, and SNP genotypes as 0, 1 or 2 depending on the number of minor alleles present. The linear effects of baseline biomarkers on PFS were assessed using a Cox proportional hazard model with the following co-variates: Eastern Cooperative Oncology Group (ECOG) performance status (0-1 versus 2), liver-only metastases (yes/no), and the number of distant metastasis organs (1 versus >1), a treatment effect, a biomarker effect and a biomarker-treatment interaction effect. The significance of the latter two effects was jointly tested by a two-degrees-of-freedom Wald test. Extended statistical methods are described in supplementary methods.

Results

(26) Of the 236 patients in the ITT population of the AFFIRM trial 227 (96%) were evaluable for response. Of these, 130 (57%) provided at least one biological sample, 60 (46%) and 70 (54%) of which participated in the mFOLFOX6 and mFOLFOX6 plus aflibercept arms, respectively. There was no major difference at a false discovery rate (FDR)-adjusted P-value of 0.05 between patients who provided a biological sample and those who did not in terms of patient biometrics, ethnicity, and disease characteristics at baseline, or at efficacy and safety endpoints (Table 1). Of those who provided at least one biological sample, 51 (39%) provided samples for each of the 3 biomarker types, with 88 (68%) and 97 (74%) patients providing samples for 2 or 1 of the biomarker types respectively. Each biomarker type was analyzed separately, to avoid patient groups that were too small for sub-analyses.

Profiling of Plasma Markers for Efficacy

(27) Plasma levels of 27 markers were measured at different time points (i.e., at baseline [87 patients]; 30 and 60 days after start of treatment [82 and 73 patients]; and 30 days after the last treatment [56 patients] as indicated on Table 5.

(28) TABLE-US-00005 TABLE 5 Number of observations per time point-total, below limit of quantification (LOQ) and of detection (LOD) EOT + 30 Baseline Day 30 Day 60 Days Total <LOQ <LOD Total <LOQ <LOD Total <LOQ <LOD Total <LOQ <LOD ANGPT1 87 1 1 80 1 0 73 0 0 56 0 0 ANGPT2 86 1 82 1 72 0 55 1 CSF2 84 28 17 80 38 16 72 32 17 54 25 16 CXCL12 86 0 0 82 2 2 72 0 0 55 0 0 FGF1 87 14 14 80 16 16 73 18 18 56 8 8 Endostatin 87 80 73 56 FGF2 87 19 11 80 20 15 73 19 11 56 10 5 FIGF 87 12 12 80 4 4 73 1 1 56 1 1 HGF 86 0 0 82 7 7 72 1 1 55 1 1 IFNG 84 6 4 80 8 3 72 6 2 54 11 8 IL10 84 0 0 80 1 1 72 2 2 54 0 0 IL12 84 3 3 80 1 1 72 3 3 54 8 7 IL1B 84 1 1 80 0 0 72 1 1 54 4 4 IL2 84 3 2 80 3 1 72 1 1 54 8 6 IL4 84 2 2 80 1 1 72 2 2 54 7 7 IL5 84 2 1 80 3 0 72 1 0 54 1 1 IL6 84 0 0 80 2 1 72 0 0 54 1 1 IL8 84 0 0 80 1 1 72 0 0 54 0 0 PGF 87 8 8 80 6 6 73 2 2 56 1 1 TNF 84 2 1 80 2 1 72 1 0 54 2 1 PDGFA 87 80 73 56 VEGFA 84 6 1 80 4 1 72 2 0 54 4 0 PDGFB 87 80 73 56 VEGFC 86 1 1 82 0 0 72 0 0 55 1 0 sFLT4 86 0 0 82 7 7 72 1 1 55 1 1 THBS2 87 80 73 56 sKDR 86 0 0 82 7 7 72 1 1 55 1 1

(29) All cytokines were measured in pg/ml, but some transformations were applied when necessary to obtain a symmetric distribution or smaller numerical values (ng/ml) for the association models, as depicted in table 6.

(30) TABLE-US-00006 TABLE 6 Selected transformations of original plasma cytokine levels (pg/ml) Cytokine Transformation ANGPT1 log ANGPT2 log CSF2 square root CXCL12 ng/ml FGF1 cubic root Endostatin log FGF2 square root FIGF cubic root HGF Log IFNG square root IL10 cubic root IL12 None IL1B cubic root IL2 square root IL4 None IL5 Log IL6 Log IL8 Log PGF ng/ml TNF square root PDGFA log VEGFA Log PDGFB log VEGFC Log sFLT4 square root(ng/ml) THBS2 log sKDR square root(ng/ml)

(31) We assessed the association of each plasma marker at baseline with PFS, while allowing for an interaction with treatment (Table 7). The lowest P-value was obtained for IL8 (P=0.0211; FDR=0.596 and P=0.0218 for interaction).

(32) TABLE-US-00007 TABLE 7 Effect of baseline plasma biomarkers on PFS. The P-values associated to the joint effect, the FDR-corrected joint effect, the plasma biomarker and the biomarker with treatment interaction effect are shown. P-values FDR- Plasma Treatment corrected level by plasma Protein Joint effect joint effect effect level effect IL8 0.0221 0.5962 0.6701 0.0218 THBS2 0.1408 0.6523 0.0545 0.2234 CXCL12 0.1610 0.6523 0.0573 0.1640 IL10 0.1669 0.6523 0.0647 0.1311 Ang1 0.1768 0.6523 0.0724 0.0831 FIGF 0.1974 0.6523 0.6826 0.1704 FGF2 0.2063 0.6523 0.1722 0.0768 sVEGFR2 0.2171 0.6523 0.1616 0.0812 PDGFA 0.2474 0.6523 0.1155 0.1052 IL6 0.2689 0.6523 0.1061 0.1838 FGF1 0.3185 0.6523 0.4759 0.1536 VEGF-A 0.3202 0.6523 0.2549 0.7133 CSF2 0.3272 0.6523 0.4278 0.1679 IL12 0.3587 0.6523 0.3664 0.1808 IFNg 0.3855 0.6523 0.8662 0.2984 IL4 0.3866 0.6523 0.7326 0.4796 PIGF 0.5008 0.7955 0.3018 0.2595 PDGFB 0.5442 0.8163 0.4795 0.2963 IL1B 0.6025 0.8439 0.3183 0.3595 HGF 0.6286 0.8439 0.3369 0.5093 IL2 0.6564 0.8439 0.7696 0.4093 Endostatin 0.9021 0.9783 0.7429 0.9824 sFLT4 0.9063 0.9783 0.6849 0.8668 TNF 0.9461 0.9783 0.8736 0.7442 Ang2 0.9698 0.9783 0.9279 0.9747 VEGF-C 0.9772 0.9783 0.8375 0.8307 IL5 0.9783 0.9783 0.8813 0.8501

(33) The Cox model assumes a linear relationship between the plasma marker and the log of the PFS hazard function, but since this hypothesis may be violated when analyzing continuous markers that vary considerably, a threshold effect may be more relevant. We explored this possibility by searching for the optimal cut-off level that maximizes the interaction with the treatment and the plasma marker. For IL8, the optimal cut-off was at 19 pg/ml (77.sup.th percentile). A model, in which IL8 was analyzed as a binary variable with this threshold, fitted better than a model with continuous IL8 levels (AIC of 469.3 versus 477.6). Patients with low IL8 levels (≥19 pg/ml, 77% of patients) exhibited a longer PFS in the aflibercept/mFLOFOX6 arm than in the mFOLFOX6 arm (Table 8).

(34) TABLE-US-00008 TABLE 8 Kaplan-Meier estimates of effect of biomarkers on months of PFS. Aflibercept plus mFOLFOX6 mFOLFOX6 Hazard Ratio Median Median vs mFOLFOX6 (99% CI) (99% CI) (99% CI) Plasma protein markers All patients 8.8 (6.57-10.02) 8.5 (6.67-10.05) 0.979 (0.505-1.897) IL8 ≤ 19 pg/mL 8.8 (5.62-10.91) 9.3 (7.52-11.10) 0.764 (0.363-1.607) IL8 > 19 pg/mL 8.8 (5.09-15.64) 4.1 (2.33-8.54) 2.71 (0.735-9.984) CI, confidence interval; mt, mutant; wt, wild-type.

(35) We also analyzed whether treatment-related changes in plasma markers could predict aflibercept treatment outcome. The Cox model included the effect of baseline plasma levels and the difference in expression measured at baseline and the last time point before discontinuation, disease progression or death (i.e., at 30 days or 60 days of treatment), while allowing for interaction with the treatment arm. IL8 was the only marker with a significant effect of change from baseline on PFS (P=0.0018; FDR=0.0478; Table 9). This effect did not differ between treatment arms (P=0.2028). High baseline or post-baseline increased IL8 levels corresponded to a higher probability of disease progression at 12 months (FIGS. 1 and 2).

(36) TABLE-US-00009 TABLE 9 Effect of plasma marker changes from baseline on PFS. P-values (and FDR) of the joint effects of plasma marker and treatment by plasma marker interactions are shown. P-values for the change from baseline and the interaction are also presented. P-values Treatment by Joint effect Change from Change from Protein Joint effect FDR-corrected baseline effect baseline effect IL8 0.0018 0.0478 0.0006 0.2028 IL10 0.0342 0.4525 0.5214 0.8204 VEGFA 0.0619 0.4525 0.0189 0.0704 CXCL12 0.0670 0.4525 0.1714 0.0318 CSF2 0.0855 0.4619 0.0266 0.1682 VEGFC 0.1127 0.5072 0.0742 0.9355 IL5 0.1886 0.7275 0.0684 0.1657 Endostatin 0.2418 0.7882 0.9988 0.2668 PDGFA 0.3092 0.7882 0.1508 0.9922 TNF 0.3209 0.7882 0.1770 0.1554 IL4 0.3211 0.7882 0.1938 0.7631 FGF2 0.3851 0.8191 0.6468 0.3957 sFLT4 0.3944 0.8191 0.2533 0.1729 THBS2 0.4677 0.8426 0.2387 0.6683 FGF1 0.4770 0.8426 0.4669 0.2242 PGF 0.4993 0.8426 0.2501 0.2755 ANGPT2 0.5638 0.8924 0.3373 0.8394 IL1B 0.6261 0.8924 0.3345 0.5414 PDGFB 0.6583 0.8924 0.3730 0.6009 IL12 0.6611 0.8924 0.7934 0.7744 IL2 0.7984 0.9482 0.9352 0.6027 sKDR 0.8375 0.9482 0.6009 0.5613 HGF 0.8383 0.9482 0.5804 0.5665 IFNG 0.8559 0.9482 0.6923 0.9101 ANGPT1 0.8814 0.9482 0.6269 0.8538 FIGF 0.9131 0.9482 0.6838 0.7528 IL6 0.9963 0.9963 0.9893 0.9594

(37) When plasma biomarkers were measured at baseline only, IL8 had the most prominent effect on PFS, which was best described as a threshold effect with high circulating IL8 (IL8>19 pg/mL) associated with a shorter PFS in the aflibercept-treated patients.

(38) When plasma biomarkers were measured at baseline and during treatment, high levels of circulating IL8 at baseline together with increased levels of IL8 measured during treatment were significantly associated with reduced PFS (FDR=0.0478).

Conclusions

(39) We identified that high IL8 levels at baseline correlated with shorter survival times, and patients with increasing levels of IL8 during treatment were more likely to progress.

(40) This suggests that patients with high IL8 levels, at baseline or during treatment, are at increased risk of disease progression during aflibercept therapy.