Abstract
BACKGROUND:
CALGB 90401 (Alliance) was a phase III trial of 1050 patients with metastatic castration-resistant prostate cancer (mCRPC) comparing docetaxel, prednisone, bevacizumab (DP+B) versus DP alone. While this trial did not show an improvement in overall survival (OS), there were improved intermediate outcomes suggesting that subsets of men may derive benefit from this combination. The purpose of this analysis was to identify prognostic and predictive biomarkers associated with OS and progression-free survival (PFS) benefit from DP+B.
METHODS:
Baseline EDTA plasma samples from 650 consenting patients were analyzed for 24 biomarkers. The proportional hazards model was utilized to test for the prognostic and predictive importance of the biomarkers for OS. The statistically significant biomarkers of OS were further investigated for prognostic and predictive importance for other secondary outcomes.
RESULTS:
15 markers [ICAM-1, VEGF-R3, TIMP-1, TSP-2, Ang-2, Her-3, Osteopontin (OPN), PlGF, VCAM-1, HGF, VEGF, Chromogranin A, IL-6, VEGF-R1, BMP-9] were prognostic of OS, while 9 markers (ICAM-1, VEGF-R3, Her-3, TIMP-1, Ang-2, OPN, PlGF, HGF, and VEGF) were also prognostic of PFS. All markers were statistically significant in univariate analyses after adjustment for multiplicity (FDR<0.1). In multivariable analyses of OS adjusting for risk score, seven markers had FDR<0.1, including ICAM-1, VEGF-R3, TIMP-1, Ang-2, VEGF, TSP-2 and HGF. In unadjusted analysis, OPN was predictive of PFS improvement with DP+B, in both univariate and multivariable analysis. However, none of the biomarkers tested were predictive of clinical outcomes after adjusting for multiple comparisons.
CONCLUSIONS:
Multiple biomarkers were identified in CALGB 90401 as prognostic of clinical outcomes but not predictive of OS. While OPN may have promise as a potential biomarker for anti-angiogenic therapies, further mechanistic and clinical studies are needed to determine the underlying biology and potential clinical application.
Keywords: bevacizumab, prognostic marker, predictive marker, prostate cancer
INTRODUCTION
In 2023 docetaxel remains the primary first line chemotherapy in men with metastatic castration-resistant prostate cancer (mCRPC), particularly after progression on potent androgen receptor (AR) inhibitors(1,2). Docetaxel was approval in 2004 by Food and Drug Administration (FDA) as the new standard of care for mCRPC, after two phase-3 trials (TAX327 and SWOG 99-16) revealed that mCRPC patients receiving docetaxel and prednisone demonstrated significantly improved overall survival (OS) and progression-free survival (PFS) (3,4). In the past decade, tremendous progress in the systemic management of prostate cancer has been made with the development of novel chemotherapy agents(5), immunotherapy(6), second generation AR inhibitors(7-9), poly (ADP-ribose) polymerase (PARP) inhibitors(10), radio-isotopes(11,12), and immune checkpoint inhibitors(13-15) as effective life-prolonging therapies for men with metastatic hormone sensitive prostate cancer (mHSPC) and mCRPC.
The development of novel drugs encouraged exploration of a multimodal strategy in advanced prostate cancer, especially because studies suggest that each modality contributes to further disease debulking and disease control(16). Vascular endothelial growth factor (VEGF) plays a pivotal role in the pathogenesis and progression of mCRPC(17). Increased VEGF plasma levels correlate with metastatic disease progression(18,19). Both plasma and urine VEGF levels are independent predictors of OS(20,21). Naturally, it was speculated that additional inhibition of VEGF may enhance current therapies in mCRPC.
The Cancer and Leukemia Group B (CALGB, the Alliance for Clinical Trials in Oncology) 90401 study evaluated the addition of bevacizumab (B) to docetaxel and prednisone (DP). While no improvement in OS in the overall population (22.6 vs 21.5 months, p=0.181) was observed, there were improvements in median PFS (9.9 vs 7.5 months; p<0.001), major prostate-specific antigen (PSA) response (69.5% vs. 57.9%; p<0.001), and objective radiologic response (49.4% vs 35.5%, p=0.0013) in patients receiving DP+B vs DP, suggesting that subsets of men may derive benefit from this drug regimen (22). The goal of the present study was to identify biomarkers that may identify men with mCRPC that have clinical benefits from the addition of bevacizumab to DP.
Predictive biomarkers enable selection of patients who are mostly likely to derive benefit from a treatment. To help identify potential biomarkers for anti-angiogenic agents, we developed and optimized the angiome, a panel of 24 circulating protein biomarkers known to play crucial roles in angiogenesis, inflammation, and immune modulation(23). We have applied this technology to multiple phase III trials and identified multiple prognostic and predictive biomarkers across various cancer types(24-28). In this retrospective correlative study, we analyzed the angiome in baseline samples from patients enrolled in CALGB 90401, and included one additional biomarker, chromogranin A (CgA), a neuroendocrine protein of specific importance in prostate cancer patients (30). Identification of prognostic or predictive biomarkers in men with mCRPC could be considered for future patient-selection or for therapeutic development of novel agents or combinations.
METHODS
Study Design and Patients
CALGB 90401 (NCT00110214) was a randomized double-blind placebo-controlled phase III trial of 1,050 mCRPC patients comparing docetaxel and prednisone (DP) with docetaxel, prednisone, and bevacizumab (DP+B) (22). Inclusion criteria included patients with progressive adenocarcinoma of the prostate despite castrate levels of testosterone, performance status of 0-2 and adequate renal and hepatic functions. Each participant in this correlative study of CALGB 90401 signed an IRB-approved informed consent document in accordance with guidelines for Good Clinical Practice and guiding principles laid out in the declaration of Helsinki. Plasma samples were collected at baseline from 650 patients who signed an IRB-approved informed consent document, in accordance with guidelines for Good Clinical Practice and the declaration of Helsinki.
Laboratory Analysis
Peripheral venous blood was collected at baseline from consenting patients and processed on-site within 2 hours. Platelet-poor EDTA plasma was aliquoted, frozen, and shipped on dry ice to the Alliance Pathology Coordinating Office for centralized storage. Aliquots of frozen plasma were shipped to the Duke Phase I Biomarker Laboratory for the angiome analyses. A total of 24 biomarkers were assessed using the multiplex enzyme-linked immunosorbent assay (ELISA) techniques, with 20 markers tested on the Quanterix (Billerica, MA) and 3 markers (BMP-9, CD73, and Her3) tested on the Meso Scale Discovery (Rockville, MD) platforms (24,25). Chromogranin A was tested using human chromogranin A Simple Step ELISA kit from Abcam (ab196271) following manufacturer’s instructions. The analyses adhered to REMARK criteria guidelines (31), and laboratory personnel were blinded to the clinical results.
Data Analysis
The primary endpoint of the biomarker analysis was overall survival (OS), defined as the time from date of random assignment to date of death or last follow-up. Secondary end points were progression-free survival (PFS) and ≥50% decline in on-treatment PSA from baseline. PFS was defined from the date of random assignment to date of biochemical, bone, soft tissue progression or death, whichever occurred first. Greater than 50% decline in PSA from baseline was defined using the PSA WG1 (32).
In all analyses, the biomarkers were considered as continuous and were modeled on a log 2 scale. The Kaplan-Meier OS and PFS distribution by low (<observed median level of a biomarker) and high levels (≥observed median level of a biomarker) were presented. The proportional hazards (PH) model was used to test the prognostic importance of each of the 24 biomarkers for OS in univariate analysis. We adjusted the p-values from the univariate analysis of OS for multiplicity using the false discovery rate (FDR)(33), considering FDR <0.1 as statistically significant. In addition, multivariable analysis adjusting for the clinical risk score (34) were performed to determine whether the statistically significant biomarkers of OS remained statistically significant with the established clinical prognostic factors. We next investigated whether the 15 statistically significant plasma biomarkers of OS were also important factors for PFS and ≥ 50% decline in PSA from baseline. For the latter analysis, we employed the logistic regression model. The power of the correlative science study and the predictive analysis are presented in the supplement.
RESULTS
Patient characteristics and biomarker baseline expression
Of the 1050 patients with mCRPC enrolled in CALGB 90401, 650 patients had plasma samples available for biomarker analysis. Of the 650 patients, 605 died during follow-up and the median follow up for the 45 surviving patients was 65.2 months (CI: 61.4-68.6). The demographic and clinical outcomes of all patients are presented in Table 1. There were no differences between the biomarker selected population and the overall population of patients treated on study. Median PFS time was 8.8 months for the biomarker selected population and 8.5 months for the overall population. There were no differences between the subgroups of patients receiving docetaxel and prednisone (DP) versus docetaxel, prednisone, and bevacizumab (DP+B) regarding age, race, ECOG status, Gleason score, etc.
Table 1. Baseline characteristics and clinical outcomes by patient groups.
| DP | DP+B | Biomarker population (n=650) |
Total (n=1050) | |
|---|---|---|---|---|
| (n=326) | (n=324) | |||
| Age at Diagnosis (years)* | 69 (62, 75) | 69 (63, 74) | 69 (63, 75) | 69 (63, 75) |
| Race % | ||||
| White | 86.5 | 87.7 | 87.1 | 87.9 |
| Black | 12.3 | 10.8 | 11.5 | 10.5 |
| Asian | 0.3 | 0.9 | 0.6 | 0.7 |
| Others | 0.9 | 0.6 | 0.8 | 1 |
| Ethnicity (%) | ||||
| Hispanic or Latino | 5.2 | 2.8 | 4 | 4.3 |
| Non-Hispanic | 89.9 | 91 | 90.5 | 90.6 |
| Unknown | 4.9 | 6.2 | 5.5 | 5.1 |
| ECOG Performance Status (%) | ||||
| 0 | 59.2 | 54.6 | 56.9 | 55.6 |
| 1 | 38.3 | 42 | 40.2 | 40.1 |
| 2 | 2.5 | 3.4 | 2.9 | 4.3 |
| Gleason Score (%) | ||||
| <8 | 50 | 51.2 | 50.6 | 50 |
| >=8 | 50 | 48.5 | 49.2 | 49.8 |
| Missing | 0 | 0.3 | 0.2 | 0.2 |
| Metastatic Site (%) | ||||
| Lymph node | 44.5 | 40.4 | 42.5 | 42.4 |
| Bone | 83.1 | 88.3 | 85.7 | 85.8 |
| Visceral | 12.6 | 19.4 | 16 | 16.2 |
| Albumin* | 4.0 (3.7, 4.2) | 4.0 (3.6, 4.3) | 4.0 (3.7, 4.2) | 4.0 (3.6, 4.2) |
| Alkaline Phosphatase* | 112 (81.0, 213) | 123 (87.8, 221) | 117 (84.0, 217) | 118 (82, 226) |
| Hemoglobin* | 12.7 (11.5, 13.8) | 12.8 (11.7, 13.8) | 12.8 (11.6, 13.8) | 12.8 (11.6, 13.8) |
| LDH* | 207 (167, 318) | 209 (169, 324) | 209 (167, 321) | 205 (166, 297) |
| Prognostic Risk Group (%) | ||||
| Low | 35.6 | 33 | 34.3 | 32.6 |
| Intermediate | 35 | 32.4 | 33.7 | 33.6 |
| High | 29.4 | 34.6 | 32 | 33.8 |
| Measurable Disease (%) | 53.4 | 48.1 | 50.8 | 49.8 |
| Median OS, months (95% CI) | 22.3 (20.2, 25.3) | 23.0 (21.1, 26.5) | 22.6 (21.3, 24.5) | 22.0 (21.1, 23.4) |
| Median PFS, months (95% CI) | 7.6 (6.7, 8.5) | 10.1 (9.2, 11.1) | 8.8 (8.1, 9.5) | 8.5 (8.0, 9.1) |
| Objective Response (% among patients with measurable disease) | N=330 | N=523 | ||
| Yes | 46.7 | 42.1 | ||
| No | N=174 | N=156 | 53.3 | 57.9 |
| 36.8 | 57.7 | |||
| 63.2 | 42.3 | |||
| PSA Decline ≥ 50% (% among patients with PSA from baseline) | N=629 | N=1007 | ||
| Yes | 66.6 | 65 | ||
| No | N=313 | N=316 | 33.4 | 35 |
| 60.4 | 72.8 | |||
| 39.6 | 27.2 |
Abbreviations: ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase; OS, overall survival; CI, confidence interval; PFS, progression-free survival.
Median (1st quartile, 3rd quartile)
Assays performed well with low coefficients of variation (CV<10%). Detailed descriptive summary for all baseline biomarkers levels (mean, median, range) is provided in Supplementary Table S1. We explored whether patients’ clinical risk scores were correlated with any specific biomarker. OPN was found to have the strongest correlation with increasing clinical risk score (Spearman correlation coefficient=0.61) while Ang-2 (0.37), IL-6 (0.34) and chromogranin A (0.33) were also found to correlate with increasing risk across the three groups (Supplementary Table S2).
Prognostic analysis of baseline biomarkers for overall survival
Fifteen of the 24 markers (ICAM-1, VEGF-R3, TIMP-1, TSP-2, Ang-2, Her-3, OPN, PlGF, VCAM-1, HGF, VEGF, Chromogranin A, IL-6, VEGF-R1, and BMP-9) were prognostic of OS (Table 2). These markers were all statistically significant in univariate analyses after adjustment for multiplicity (FDR<0.1). In almost all cases, shorter OS times were associated with higher than median levels for all markers, with the only exception being BMP-9 (OS of 21.3 vs. 25.1 months in marker low vs. high groups, HR 0.9; 95% CI= 0.8-1.0). In multivariable analyses of OS adjusting for risk score, seven markers had FDR<0.1, including ICAM-1 (HR=1.5, 95% CI=1.2-1.9), VEGF-R3 (HR=1.4, 95% CI=1.2-1.7), TIMP-1 (HR=1.3, 95% CI=1.1-1.5), Ang-2 (HR=1.2, 95% CI=1.1-1.3), VEGF (HR=1.1, 95% CI=1.0-1.2), TSP-2 (HR=1.2, 95% CI=1.0-1.3) and HGF (HR=1.1, 95% CI=1.0-1.2; Table 2). It should be noted that given the association of OPN with both clinical risk score and OS, the multivariable analysis of OPN may be confounded due to its potential role in the underlying biology of this disease. A list of prognostic value for OS for the remaining biomarkers is presented in Supplemental Table S3.
Table 2. Prognostic markers for overall survival (OS).
Median OS by dichotomized median marker, univariate and multivariable hazard ratios (HRs) from the proportional hazards models of plasma markers (modeled on a log2 scale) predicting OS. Ordered by hazard ratios from univariate analysis.
| Biomarker | Median OS (months, 95% CI) | Univariate | Multivariable** | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Low | High | N | HR (95% CI) | P Value | FDR* | HR (95% CI) | P Value | FDR* | |
| ICAM-1 | 26.81 (24.21-29.63) | 20.07 (18.40-22.14) | 649 | 1.87 (1.49-2.34) | 0 | 0 | 1.49 (1.20-1.86) | 0 | 0.002 |
| VEGF-R3 | 27.86 (25.23-31.44) | 19.58 (18.17-21.62) | 649 | 1.72 (1.41-2.09) | 0 | 0 | 1.42 (1.16-1.73) | 0.001 | 0.002 |
| TIMP-1 | 26.97 (24.31-28.85) | 19.91 (18.17-21.91) | 649 | 1.56 (1.35-1.80) | 0 | 0 | 1.30 (1.13-1.49) | 0 | 0.002 |
| TSP-2 | 26.81 (24.51-29.31) | 20.01 (18.17-21.42) | 649 | 1.46 (1.30-1.64) | 0 | 0 | 1.15 (1.02-1.30) | 0.022 | 0.056 |
| Ang-2 | 28.25 (26.51-32.16) | 18.76 (16.99-20.67) | 650 | 1.39 (1.27-1.53) | 0 | 0 | 1.19 (1.08-1.31) | 0 | 0.002 |
| Her-3 | 25.17 (21.82-27.83) | 21.45 (20.01-23.43) | 648 | 1.34 (1.03-1.75) | 0.028 | 0.052 | 1.17 (0.90-1.52) | 0.233 | 0.323 |
| OPN | 28.45 (26.51-31.70) | 18.50 (16.26-20.07) | 649 | 1.31 (1.23-1.40) | 0 | 0 | 1.06 (0.98-1.14) | 0.149 | 0.251 |
| PlGF | 26.78 (23.36-29.11) | 20.86 (19.29-22.67) | 649 | 1.27 (1.13-1.42) | 0 | 0 | 1.09 (0.97-1.22) | 0.151 | 0.251 |
| VCAM-1 | 24.15 (22.34-26.74) | 20.21 (18.40-23.95) | 649 | 1.21 (1.00-1.47) | 0.044 | 0.072 | 1.05 (0.87-1.26) | 0.625 | 0.721 |
| HGF | 25.63 (23.29-27.89) | 20.07 (18.46-22.60) | 649 | 1.18 (1.09-1.27) | 0 | 0 | 1.08 (1.00-1.17) | 0.043 | 0.092 |
| VEGF | 25.63 (21.88-28.25) | 21.45 (19.29-23.43) | 649 | 1.15 (1.07-1.24) | 0 | 0 | 1.10 (1.03-1.18) | 0.008 | 0.023 |
| CgA # | 26.51 (23.69-28.25) | 19.68 (16.66-21.91) | 648 | 1.13 (1.07-1.19) | 0 | 0 | 1.03 (0.98-1.09) | 0.259 | 0.323 |
| IL-6 | 25.63 (23.43-28.81) | 19.98 (18.37-21.91) | 650 | 1.07 (1.03-1.11) | 0 | 0.001 | 1.00 (0.96-1.04) | 0.855 | 0.916 |
| VEGF-R1 | 24.02 (21.98-26.97) | 21.42 (19.12-24.15) | 650 | 1.06 (1.00-1.13) | 0.045 | 0.072 | 1.00 (0.94-1.06) | 0.99 | 0.99 |
| BMP-9 | 21.32 (19.91-22.83) | 25.13 (22.31-27.86) | 648 | 0.91 (0.84-0.98) | 0.016 | 0.032 | 0.95 (0.88-1.04) | 0.258 | 0.323 |
Abbreviation: FDR, false discovery rate.
CgA: chromogranin A
adjusting for 24 biomarkers
adjusting for clinical risk score
95% CI were not adjusted for multiplicity, thus these intervals should not be used to infer definitive relationships with clinical outcome
Prognostic analysis of baseline markers for progression-free survival
In subsequent univariate analyses, we examined the 15 statistically significant biomarkers for OS and 9 of these biomarkers (ICAM-1, VEGF-R3, Her-3, TIMP-1, Ang-2, OPN, PlGF, HGF, VEGF) were also prognostic of PFS with FDR <0.1 in univariate analysis (Table 3). All nine biomarkers had estimated HR>1, indicating that higher biomarker levels were associated with worse PFS. Patients with low levels of the biomarkers (less than the observed median levels) had an average of 1.7 month longer median PFS than patients will high levels of the biomarkers. None of the biomarkers were statistically significant of PFS in the multivariable analayses when adjusting for clinical risk score and treatment arm. A list of the prognostic value for PFS for the remaining biomarkers is shown in Supplemental Table S3.
Table 3. Prognostic markers for progression-free survival (PFS).
Median PFS by dichotomized median biomarker, univariate and multivariable hazard ratios (HRs) from the proportional hazards models of plasma markers (modeled on a log2 scale) predicting PFS. Ordered by the hazard ratios from the univariate analysis.
| Biomarker | Median PFS (months, 95% CI) |
Univariate | Multivariable** | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Low | High | N | HR (95% CI) | P Value | FDR* | HR (95% CI) | P Value | FDR* | |
| ICAM-1 | 9.43 (8.67-10.58) | 8.08 (7.23- 9.10) | 649 | 1.40 (1.12-1.75) | 0.003 | 0.011 | 1.25 (1.00-1.55) | 0.048 | 0.18 |
| VEGF-R3 | 9.76 (8.97-10.51) | 7.75 (6.90- 8.38) | 649 | 1.39 (1.14-1.68) | 0.001 | 0.006 | 1.23 (1.01-1.50) | 0.037 | 0.18 |
| Her-3 | 9.76 (8.77-10.68) | 7.95 (7.56- 8.87) | 648 | 1.35 (1.05-1.73) | 0.02 | 0.052 | 1.29 (1.01-1.65) | 0.044 | 0.18 |
| TIMP-1 | 9.26 (8.48-10.18) | 8.08 (7.49- 9.26) | 649 | 1.33 (1.16-1.53) | 0 | 0 | 1.19 (1.03-1.37) | 0.015 | 0.18 |
| Ang-2 | 9.82 (8.61-10.74) | 7.85 (7.23- 8.97) | 650 | 1.18 (1.08-1.29) | 0 | 0.002 | 1.07 (0.98-1.18) | 0.137 | 0.294 |
| OPN | 10.25 (9.69-11.30) | 7.23 (6.18- 7.95) | 649 | 1.16 (1.09-1.23) | 0 | 0 | 1.05 (0.97-1.13) | 0.239 | 0.403 |
| PlGF | 9.53 (8.67-10.48) | 8.05 (7.56- 8.97) | 649 | 1.14 (1.03-1.27) | 0.012 | 0.037 | 1.07 (0.96-1.19) | 0.242 | 0.403 |
| HGF | 9.56 (8.51-10.81) | 8.02 (7.46- 9.10) | 649 | 1.14 (1.05-1.23) | 0.002 | 0.007 | 1.08 (0.99-1.17) | 0.074 | 0.204 |
| TSP-2 | 9.66 (8.94-10.48) | 7.75 (7.23- 8.87) | 649 | 1.12 (1.00-1.25) | 0.042 | 0.101 | 0.98 (0.87-1.11) | 0.786 | 0.842 |
| VEGF | 9.43 (8.67-10.51) | 7.95 (7.56- 8.97) | 649 | 1.10 (1.02-1.18) | 0.009 | 0.03 | 1.07 (0.99-1.15) | 0.081 | 0.204 |
| VCAM-1 | 9.53 (8.48-10.25) | 8.08 (7.43- 9.10) | 649 | 1.05 (0.87-1.26) | 0.616 | 0.672 | 0.96 (0.80-1.16) | 0.698 | 0.805 |
| CgA # | 9.36 (8.38-10.05) | 8.08 (7.59- 9.10) | 648 | 1.05 (1.00-1.11) | 0.059 | 0.13 | 1.00 (0.95-1.06) | 0.945 | 0.945 |
| IL-6 | 9.63 (8.87-10.25) | 7.85 (7.13- 8.84) | 650 | 1.02 (0.99-1.06) | 0.195 | 0.276 | 0.98 (0.94-1.02) | 0.319 | 0.478 |
| VEGF-R1 | 9.40 (8.38-10.25) | 8.08 (7.69- 9.13) | 650 | 1.02 (0.96-1.08) | 0.483 | 0.58 | 0.98 (0.92-1.04) | 0.486 | 0.663 |
| BMP-9 | 8.51 (7.95- 9.40) | 9.10 (8.08-10.05) | 648 | 0.99 (0.91-1.06) | 0.706 | 0.737 | 1.02 (0.95-1.10) | 0.596 | 0.745 |
adjusting for 15 biomarkers
adjusting on clinical risk score and treatment arm
95% CI were not adjusted for multiplicity, thus these intervals should not be used to infer definitive relationships with clinical outcome
Predictive analysis with baseline markers and treatment groups
We next explored whether any of the 24 biomarkers were predictive of bevacizumab benefit when added to docetaxel chemotherapy. Using the PH model that included treatment interaction term (intx), three biomarkers appeared to be predictive of OS improvement in unadjusted analysis with p values close to significance, consisting of OPN (Pintx=0.063), PDGF-BB (Pintx=0.064), and PDGF-AA (Pintx=0.094) (Table 4). The median OS for low OPN patients was 28.5 (95% CI: 25.2-31.9) months in the DP arm and 28.5 (95% CI: 25.1-33.0) months in the DP+B arm. However, median OS improved from 17.4 (95% CI: 14.1-20.3) to 19.7 months (95% CI: 16.3-21.8) in men with high OPN levels upon addition of B to DP. In multivariable analyses where we adjusted for clinical risk score and multiplicity, OPN and PDGF-BB were not predictive of OS (Table 4). A list of predictive value for the remaining biomarkers for OS is shown in Supplemental Table S4.
Table 4. Predictive markers of overall survival (OS) and progression-free survival (PFS).
Median and 95% confidence intervals (CIs) for OSPFS and PFSOS by treatment and dichotomized by biomarker median level. Ordered by the univariate hazard ratios of the low-level group in OS.
| Biomarker | Level | Median OS (months, 95% CI) | Model | Univariate Analysis | Multivariable Analysis*** | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| DP | DP+B | HR (95% CI) | Pintx* | FDR** | HR (95% CI) | Pintx* | FDR** | |||
| OPN | Low | 28.45 (25.17-31.93) | 28.53 (25.13-32.95) | OPN | 1.4 (1.28, 1.53) | 1.13 (1.02, 1.25) | ||||
| High | 17.45 (14.13-20.34) | 19.75 (16.33-21.82) | Arm | 10.28 (0.88, 119.87) | 9.2 (0.9, 94.18) | |||||
| OPN*Arm | 0.89 (0.79, 1.01) | 0.063 | 0.496 | 0.89 (0.8, 1) | 0.054 | 0.66 | ||||
| Clinical Risk Score | NA | 2.95 (2.28, 3.81) | ||||||||
| PDGF-AA | Low | 21.06 (18.60-26.12) | 23.36 (20.76-26.97) | PDGF-AA | 0.90 (0.81, 1.00) | 0.95 (0.86, 1.06) | ||||
| High | 23.29 (21.06-27.47) | 22.75 (20.07-27.56) | Arm | 0.26 (0.05, 1.32) | 0.31 (0.07, 1.45) | |||||
| PDGF-AA *Arm | 1.13 (0.98, 1.32) | 0.094 | 0.496 | 1.11 (0.96, 1.28) | 0.158 | 0.687 | ||||
| Clinical Risk Score | NA | 3.33 (2.73, 4.04) | ||||||||
| PDGF-BB | Low | 23.13 (20.17-27.76) | 26.64 (22.51-31.34) | PDGF-BB | 0.92 (0.81, 1.05) | 0.95 (0.83, 1.08 | ||||
| High | 21.91 (19.55-25.17) | 21.14 (19.75-24.21) | Arm | 0.14 (0.02, 1.17) | 0.13 (0.02, 0.99) | |||||
| PDGF-BB *Arm | 1.19 (0.99, 1.44) | 0.064 | 0.496 | 1.19 (1, 1.42) | 0.055 | 0.66 | ||||
| Clinical Risk Score | NA | 3.35 (2.76, 4.07) | ||||||||
| Biomarker | Level | Median PFS (months, 95% CI) | Model | Univariate Analysis | Multivariable Analysis*** | |||||
| DP | DP+B | Pintx* | FDR** | HR (95% CI) | Pintx* | FDR** | ||||
| OPN | Low | 9.40 (8.25-10.51) | 11.24 (10.05-12.62) | OPN | 1.27 (1.16, 1.39) | 1.13 (1.02, 1.25) | ||||
| High | 5.95 (5.09- 7.23) | 8.31 (7.26-10.18) | Arm | 14.94 (1.32, 168.74) | 10.9 (1.01, 117.79) | |||||
| OPN*Arm | 0.86 (0.76, 0.97) | 0.016 | 0.324 | 0.88 (0.78, 0.99) | 0.027 | 0.224 | ||||
| Clinical Risk Score | NA | 1.69 (1.32, 2.17) | ||||||||
| PDGF-AA | Low | 7.62 (6.51-9.10) | 9.76 (8.51-10.81) | PDGF-AA | 1.01 (0.91, 1.12) | 1.05 (0.95, 1.17) | ||||
| High | 7.59 (6.44-8.71) | 10.78 (9.63-11.86) | Arm | 0.59 (0.12, 2.94) | 0.84 (0.17, 4.08) | |||||
| PDGF-AA *Arm | 1.03 (0.89, 1.19) | 0.722 | 0.866 | 0.99 (0.86, 1.14) | 0.879 | 0.902 | ||||
| Clinical Risk Score | NA | 1.91 (1.58, 2.31) | ||||||||
| PDGF-BB | Low | 7.85 (6.54-9.26) | 10.25 (9.23-12.22) | PDGF-BB | 1.01 (0.88, 1.15) | 1.03 (0.9, 1.17) | ||||
| High | 7.56 (6.31-8.48) | 9.79 (8.31-11.14) | Arm | 0.19 (0.02, 1.64) | 0.27 (0.03, 2.26) | |||||
| PDGF-BB *Arm | 1.13 (0.94, 1.37) | 0.192 | 0.487 | 1.09 (0.91, 1.32) | 0.342 | 0.628 | ||||
| Clinical Risk Score | NA | 1.88 (1.56, 2.27) | ||||||||
Abbreviations: DP, docetaxel and prednisone; DP+B, docetaxel and prednisone plus bevacizumab.
p value for the interaction term (Marker * Treatment Arm). (model: Survival = Marker (continuous, log2) + Treatment + Marker*Treatment)
adjusting for 24 biomarkers
multivariable model adjusting on clinical risk score
95% CI were not adjusted for multiplicity, thus these intervals should not be used to infer definitive relationships with clinical outcome
In unadjusted analysis, OPN was predictive of PFS improvement with DP+B, in both univariate (Pintx=0.016), and multivariable analysis (Pintx=0.027), (Table 4 and Figure 1). The median PFS in low OPN patients improved from 9.4 months (95% CI: 8.2-10.5) to 11.2 months (95% CI: 10.1-12.6) with the addition of B to DP; in high OPN patients, the median PFS improved from 5.9 (95% CI: 5.1-7.2) to 8.3 (95% CI: 7.3-10.2) with the addition of B to DP (Figure 1). Kaplan-Meier plots for the rest of 23 markers for OS and PFS were shown in Supplemental Figures.
Figure 1.
Kaplan-Meier curves of OPN and treatment for overall survival (A) curves and progression free survival (B). For illustrative purpose, OPN levels were dichotomized relative to the median level. Patients with higher than median OPN level derived more benefit from DP+B, with interaction P values of 0.063 for OS and 0.016 for PFS.
DISCUSSION
VEGF was historically identified as a target in men with mCRPC based on preclinical studies linking angiogenesis and VEGF levels with disease aggressiveness and the clear prognostic role of VEGF levels with outcome (3,17). After encouraging phase II results were reported showing considerable efficacy over historic data with docetaxel alone, the randomized phase III CALGB 90401 trial evaluating docetaxel/prednisone with or without bevacizumab in men with mCRPC was conducted. While the trial did not meet its primary objective of extending OS, a significant improvement in PFS and response rates suggested that inhibiting VEGF signaling during docetaxel chemotherapy might be beneficial in some patients. The discrepancy between OS and other intermediate outcomes was likely caused by several factors, such as inadequate patient selection, discontinuation of bevacizumab after progression, and the management of treatment-related toxicities, including increased cardiovascular toxicity and infection risk due to the addition of bevacizumab.
To improve therapeutic benefit, patient selection remains pivotal for future drug development and treatment optimization. To this end, we evaluated the angiome biomarkers in baseline plasma samples collected from 650 patients in CALGB 90401 and identified several key findings. First, we identified a consistent set of prognostic biomarkers for both OS and PFS. While the magnitude of the effects differed, the direction of each effect remained consistent across both OS and PFS endpoints. Functionally, these nine overlapping prognostic markers can be grouped into angiogenic factors (Ang-2, HGF, PlGF, VEGF, and VEGF-R3), tumor microenvironment modulating factors (OPN, TIMP-1, ICAM-1) and Her-3, a soluble form of the receptor tyrosine-protein kinase erbB-3(35). Higher expression of all markers correlated with worse outcomes. Notably, all nine markers prognostic for PFS were also prognostic for OS, while six biomarkers were prognostic for OS only. These six OS prognostic markers included chromogranin A, a neuroendocrine lineage plasticity factor previously implicated as a poor prognostic biomarker in men with mCRPC(30). Of all these biomarkers, only osteopontin (OPN) was consistently associated with both baseline clinical risk score, OS, and PFS, suggesting its potential functional relevance in causing aggressive disease biology or a strong association with disease burden.
Second, in unadjusted, univariate analyses, we identified three putative predictive biomarkers for OS consisting of OPN, PDGF-AA, and PDGF-BB; OPN was also associated with improved PFS upon the addition of bevacizumab. None of these biomarkers were significant after multiple comparison testing. However, we did observe that patients with higher OPN expression generally experienced worse outcomes (negative prognostic factor), and this was the population of patients who derived significant delays in progression and improved survival from the addition of bevacizumab to DP (Figure 1). Given the importance of PSA monitoring in prostate cancer progression (36), we assessed potential prognostic/predictive value of the biomarkers in an explorative manner. OPN was one of the eight markers significantly associated with this secondary clinical outcome (Supplemental Table S5). Lower OPN level is associated with higher likelihood of PSA decline, consistent with its negative prognostic role in the OS/PFS analysis.
Interestingly, we have also observed that OPN predicts for PFS benefit in colorectal cancer patients randomized to receive chemotherapy with regorafenib versus chemotherapy alone (23). Though defining the mechanism of why patients with high OPN levels derive more benefit from bevacizumab is beyond the scope of this study, we speculate that OPN plays pivotal roles in chemoresistance(37) and immune suppression(38-40). Tumor associated macrophages (TAM) have been identified as key inducers of the angiogenic switch after exposure to bevacizumab. OPN is known to be secreted by TAMs and been suggested to promote chemoresistance and tumor recurrence. Both OPN and VEGF are co-secreted by TAMs, patients with high OPN levels may be more heavily dependent on angiogenic processes, thus more sensitive to bevacizumab.
Our biomarker findings are generally in agreement with the literature. In the randomized phase II trial of androgen deprivation therapy with or without bevacizumab in men with PC, higher levels of OPN or IL-6 were associated with increased benefit from bevacizumab(41). The original analysis of CALGB 9480, a phase III trial where patients were randomized to three levels of suramin (a chemical with no known anti-angiogenic functions), initially identified urine VEGF as a predictive marker (20); IL-6 (21), chromogranin A (30), and HGF (42) were identified as prognostic factors of OS in mCRPC patients. Here the prognostic value of VEGF, IL-6, Chromogranin A, and HGF was validated in the larger, randomized trial of CALGB 90401.
Compared to most biomarker research where often only a few biomarkers are evaluated in an inconsistent manner, our optimized multiplex angiome has been consistently applied to clinical trial samples, providing a comprehensive and informative set of biomarker analyses. However, this study is not without limitations. Three biomarkers were identified as predictive for OS benefit, but only one biomarker (OPN) was associated with both OS and PFS benefits. OPN was prognostic of OS and PFS, but the association was lost after adjustment for clinical factors, suggesting that OPN may be either an indicator of aggressive disease and tumor burden, or a driver of this aggressive disease and tumor burden. Certain therapies are known to improve OS without providing PFS benefit, while other therapies modestly delay PFS without impacting OS, primarily due to toxicities or post-progression acceleration of disease.
In summary, while we have identified several biomarkers that were prognostic of OS and PFS, none of them were predictive of OS after adjusting for clinical risk scores and multiple comparisons. Nevertheless, OPN was identified as a potential predictive biomarker of the PFS benefits of bevacizumab in addition to docetaxel/prednisone in men with mCRPC. Further study is needed to better understand the role of OPN biology in the context of anti-angiogenic therapy.
Supplementary Material
ACKNOWLEDGEMENTS
We gratefully acknowledge the invaluable contributions of the patients, their families, and the staff who participated in this study. The authors would also like to thank the Alliance Biorepository at the Ohio State University and administrative staff. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of National Cancer Institute. ClinicalTrials.gov Identifier: NCT00110214
FUNDING
Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Numbers U10CA180821, U10CA180882, and U24CA196171 (to the Alliance for Clinical Trials in Oncology), UG1CA233180, UG1CA233253, and UG1CA233341. This research is supported by the United States Army Medical Research W81XWH-18-1-0278 and the Prostate Cancer Foundation Challenge Award. Detailed information is listed in https://acknowledgments.alliancefound.org. This work is also supported in part by funds from Genentech. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
CONFLICTS OF INTEREST
A.B. Nixon has received research funding from Genentech, HTG Molecular Diagnostics, MedImmune/AstraZeneca, Medpacto, Promega Corporation, Seattle Genetics; and has received consultant/advisory compensation from AdjuVolt Therapeutics, Eli Lilly, GSK, Leap Therapeutics, Promega Corporation. D.J. George has received consultant/advisory compensation from Astellas, Astrazeneca, Axess Oncology, Bayer H/C Pharmaceuticals, BMS, Capio Biosciences, Constellation Pharmaceuticals, EMD Serono, Exelixis Inc., Flatiron, Ipsen, Janssen Pharmaceuticals, Merck Sharp & Dohme, Michael J Hennessey Associates, Modra Pharmaceuticals B. V., Myovant Sciences, Nektar Therapeutics, Physician Education Resource LLC, Pfizer, Propella TX, RevHealth LLC, Sanofi, UroGPO; and has received research funding from Astrazeneca, BMS, Calithera, Exelixis Inc., Janssen Pharmaceuticals, Novartis, Pfizer, Sanofi. H. Beltran has served as consultant/advisory board member for Janssen, Sanofi Genzyme, Astellas, Astra Zeneca, Merck, Pfizer, Foundation Medicine, Blue Earth Diagnostics, Amgen, Oncorus and has received research funding (to institution) from Janssen Oncology, AbbVie/Stemcentrx, Eli Lilly, Millennium Pharmaceuticals, Bristol Myers Squibb. A.J. Armstrong is a paid consultant with Pfizer, Astellas, Forma, BMS, Janssen, Bayer, Astrazeneca, Novartis, and Merck and receives research funding (to his institution) from Pfizer, Astellas, Janssen, Bayer, Dendreon, Novartis, Genentech/Roche, Merck, BMS, Astrazeneca, Constellation, Beigene, Forma, and Amgen. S. Halabi was on the DMC for Aveo, BMS, Janssen and Sanofi and receives research funding from Astellas and ASCO (Institution). The other authors declare no potential conflicts of interest.
DATA AVAILABILITY
The clinical data are available from the NCI data Archive. The laboratory data are available from the first author.
REFERENCES
- 1.Teo MY, Rathkopf DE, Kantoff P. Treatment of Advanced Prostate Cancer. Annu Rev Med 2019;70:479–99 doi 10.1146/annurev-med-051517-011947. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gourdin T. Recent progress in treating advanced prostate cancer. Curr Opin Oncol 2020;32(3):210–5 doi 10.1097/CCO.0000000000000624. [DOI] [PubMed] [Google Scholar]
- 3.Petrylak DP, Tangen CM, Hussain MH, Lara PN Jr., Jones JA, Taplin ME, et al. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004;351(15):1513–20 doi 10.1056/NEJMoa041318. [DOI] [PubMed] [Google Scholar]
- 4.Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi KN, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004;351(15):1502–12 doi 10.1056/NEJMoa040720. [DOI] [PubMed] [Google Scholar]
- 5.de Bono JS, Oudard S, Ozguroglu M, Hansen S, Machiels JP, Kocak I, et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet 2010;376(9747):1147–54 doi 10.1016/S0140-6736(10)61389-X. [DOI] [PubMed] [Google Scholar]
- 6.Kantoff PW, Higano CS, Shore ND, Berger ER, Small EJ, Penson DF, et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010;363(5):411–22 doi 10.1056/NEJMoa1001294. [DOI] [PubMed] [Google Scholar]
- 7.Smith MR, Saad F, Chowdhury S, Oudard S, Hadaschik BA, Graff JN, et al. Apalutamide Treatment and Metastasis-free Survival in Prostate Cancer. N Engl J Med 2018;378(15):1408–18 doi 10.1056/NEJMoa1715546. [DOI] [PubMed] [Google Scholar]
- 8.Beer TM, Armstrong AJ, Rathkopf DE, Loriot Y, Sternberg CN, Higano CS, et al. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med 2014;371(5):424–33 doi 10.1056/NEJMoa1405095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller K, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med 2012;367(13):1187–97 doi 10.1056/NEJMoa1207506. [DOI] [PubMed] [Google Scholar]
- 10.Mateo J, Carreira S, Sandhu S, Miranda S, Mossop H, Perez-Lopez R, et al. DNA-Repair Defects and Olaparib in Metastatic Prostate Cancer. N Engl J Med 2015;373(18):1697–708 doi 10.1056/NEJMoa1506859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Parker C, Nilsson S, Heinrich D, Helle SI, O'Sullivan JM, Fossa SD, et al. Alpha emitter radium-223 and survival in metastatic prostate cancer. N Engl J Med 2013;369(3):213–23 doi 10.1056/NEJMoa1213755. [DOI] [PubMed] [Google Scholar]
- 12.McBean R, O'Kane B, Parsons R, Wong D. Lu177-PSMA therapy for men with advanced prostate cancer: Initial 18 months experience at a single Australian tertiary institution. J Med Imaging Radiat Oncol 2019;63(4):538–45 doi 10.1111/1754-9485.12891. [DOI] [PubMed] [Google Scholar]
- 13.Beer TM, Kwon ED, Drake CG, Fizazi K, Logothetis C, Gravis G, et al. Randomized, Double-Blind, Phase III Trial of Ipilimumab Versus Placebo in Asymptomatic or Minimally Symptomatic Patients With Metastatic Chemotherapy-Naive Castration-Resistant Prostate Cancer. J Clin Oncol 2017;35(1):40–7 doi 10.1200/JCO.2016.69.1584. [DOI] [PubMed] [Google Scholar]
- 14.Graff JN, Alumkal JJ, Drake CG, Thomas GV, Redmond WL, Farhad M, et al. Early evidence of anti-PD-1 activity in enzalutamide-resistant prostate cancer. Oncotarget 2016;7(33):52810–7 doi 10.18632/oncotarget.10547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Antonarakis ES, Piulats JM, Gross-Goupil M, Goh J, Ojamaa K, Hoimes CJ, et al. Pembrolizumab for Treatment-Refractory Metastatic Castration-Resistant Prostate Cancer: Multicohort, Open-Label Phase II KEYNOTE-199 Study. J Clin Oncol 2020;38(5) doi 10.1200/JCO.19.01638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.O'Shaughnessy MJ, McBride SM, Vargas HA, Touijer KA, Morris MJ, Danila DC, et al. A Pilot Study of a Multimodal Treatment Paradigm to Accelerate Drug Evaluations in Early-stage Metastatic Prostate Cancer. Urology 2017;102:164–72 doi 10.1016/j.urology.2016.10.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ferrer FA, Miller LJ, Lindquist R, Kowalczyk P, Laudone VP, Albertsen PC, et al. Expression of vascular endothelial growth factor receptors in human prostate cancer. Urology 1999;54(3):567–72 doi 10.1016/s0090-4295(99)00156-9. [DOI] [PubMed] [Google Scholar]
- 18.Duque JL, Loughlin KR, Adam RM, Kantoff PW, Zurakowski D, Freeman MR. Plasma levels of vascular endothelial growth factor are increased in patients with metastatic prostate cancer. Urology 1999;54(3):523–7 doi 10.1016/s0090-4295(99)00167-3. [DOI] [PubMed] [Google Scholar]
- 19.Duque JL, Loughlin KR, Adam RM, Kantoff P, Mazzucchi E, Freeman MR. Measurement of plasma levels of vascular endothelial growth factor in prostate cancer patients: relationship with clinical stage, Gleason score, prostate volume, and serum prostate-specific antigen. Clinics (Sao Paulo) 2006;61(5):401–8 doi 10.1590/s1807-59322006000500006. [DOI] [PubMed] [Google Scholar]
- 20.Bok RA, Halabi S, Fei DT, Rodriquez CR, Hayes DF, Vogelzang NJ, et al. Vascular endothelial growth factor and basic fibroblast growth factor urine levels as predictors of outcome in hormone-refractory prostate cancer patients: a cancer and leukemia group B study. Cancer Res 2001;61(6):2533–6. [PubMed] [Google Scholar]
- 21.George DJ, Halabi S, Shepard TF, Sanford B, Vogelzang NJ, Small EJ, et al. The prognostic significance of plasma interleukin-6 levels in patients with metastatic hormone-refractory prostate cancer: results from cancer and leukemia group B 9480. Clin Cancer Res 2005;11(5):1815–20 doi 10.1158/1078-0432.CCR-04-1560. [DOI] [PubMed] [Google Scholar]
- 22.Kelly WK, Halabi S, Carducci M, George D, Mahoney JF, Stadler WM, et al. Randomized, double-blind, placebo-controlled phase III trial comparing docetaxel and prednisone with or without bevacizumab in men with metastatic castration-resistant prostate cancer: CALGB 90401. J Clin Oncol 2012;30(13):1534–40 doi 10.1200/JCO.2011.39.4767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Liu Y, Lyu J, Bell Burdett K, Sibley AB, Hatch AJ, Starr MD, et al. Prognostic and Predictive Biomarkers in Patients with Metastatic Colorectal Cancer Receiving Regorafenib. Mol Cancer Ther 2020;19(10):2146–54 doi 10.1158/1535-7163.MCT-20-0249. [DOI] [PubMed] [Google Scholar]
- 24.Nixon AB, Pang H, Starr MD, Friedman PN, Bertagnolli MM, Kindler HL, et al. Prognostic and predictive blood-based biomarkers in patients with advanced pancreatic cancer: results from CALGB80303 (Alliance). Clin Cancer Res 2013;19(24):6957–66 doi 10.1158/1078-0432.CCR-13-0926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hatch AJ, Sibley AB, Starr MD, Brady JC, Jiang C, Jia J, et al. Blood-based markers of efficacy and resistance to cetuximab treatment in metastatic colorectal cancer: results from CALGB 80203 (Alliance). Cancer Med 2016;5(9):2249–60 doi 10.1002/cam4.806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Alvarez Secord A, Bell Burdett K, Owzar K, Tritchler D, Sibley AB, Liu Y, et al. Predictive Blood-Based Biomarkers in Patients with Epithelial Ovarian Cancer Treated with Carboplatin and Paclitaxel with or without Bevacizumab: Results from GOG-0218. Clin Cancer Res 2020;26(6):1288–96 doi 10.1158/1078-0432.CCR-19-0226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Nixon AB, Sibley AB, Liu Y, Hatch AJ, Jiang C, Mulkey F, et al. Plasma Protein Biomarkers in Advanced or Metastatic Colorectal Cancer Patients Receiving Chemotherapy With Bevacizumab or Cetuximab: Results from CALGB 80405 (Alliance). Clin Cancer Res 2022;28(13):2779–88 doi 10.1158/1078-0432.CCR-21-2389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nixon AB, Halabi S, Liu Y, Starr MD, Brady JC, Shterev I, et al. Predictive Biomarkers of Overall Survival in Patients with Metastatic Renal Cell Carcinoma Treated with IFNalpha +/− Bevacizumab: Results from CALGB 90206 (Alliance). Clin Cancer Res 2022;28(13):2771–8 doi 10.1158/1078-0432.CCR-21-2386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Armstrong AJ, Nixon AB, Carmack A, Yang Q, Eisen T, Stadler WM, et al. Angiokines Associated with Targeted Therapy Outcomes in Patients with Non-Clear Cell Renal Cell Carcinoma. Clin Cancer Res 2021;27(12):3317–28 doi 10.1158/1078-0432.CCR-20-4504. [DOI] [PubMed] [Google Scholar]
- 30.Taplin ME, George DJ, Halabi S, Sanford B, Febbo PG, Hennessy KT, et al. Prognostic significance of plasma chromogranin a levels in patients with hormone-refractory prostate cancer treated in Cancer and Leukemia Group B 9480 study. Urology 2005;66(2):386–91 doi 10.1016/j.urology.2005.03.040. [DOI] [PubMed] [Google Scholar]
- 31.Altman DG, McShane LM, Sauerbrei W, Taube SE. Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK): explanation and elaboration. PLoS Med 2012;9(5):e1001216 doi 10.1371/journal.pmed.1001216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Bubley GJ, Carducci M, Dahut W, Dawson N, Daliani D, Eisenberger M, et al. Eligibility and response guidelines for phase II clinical trials in androgen-independent prostate cancer: recommendations from the Prostate-Specific Antigen Working Group. J Clin Oncol 1999;17(11):3461–7 doi 10.1200/JCO.1999.17.11.3461. [DOI] [PubMed] [Google Scholar]
- 33.Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I. Controlling the false discovery rate in behavior genetics research. Behav Brain Res 2001;125(1-2):279–84 doi 10.1016/s0166-4328(01)00297-2. [DOI] [PubMed] [Google Scholar]
- 34.Halabi S, Lin CY, Kelly WK, Fizazi KS, Moul JW, Kaplan EB, et al. Updated prognostic model for predicting overall survival in first-line chemotherapy for patients with metastatic castration-resistant prostate cancer. J Clin Oncol 2014;32(7):671–7 doi 10.1200/JCO.2013.52.3696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Li Q, Yuan Z, Cao B. The function of human epidermal growth factor receptor-3 and its role in tumors (Review). Oncol Rep 2013;30(6):2563–70 doi 10.3892/or.2013.2754. [DOI] [PubMed] [Google Scholar]
- 36.Lunger L, Tauber R, Feuerecker B, Gschwend JE, Eiber M, Heck MM. Narrative review: prostate-specific membrane antigen-radioligand therapy in metastatic castration-resistant prostate cancer. Transl Androl Urol 2021;10(10):3963–71 doi 10.21037/tau-20-1135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mirzaei A, Mohammadi S, Ghaffari SH, Yaghmaie M, Vaezi M, Alimoghaddam K, et al. Osteopontin b and c Splice isoforms in Leukemias and Solid Tumors: Angiogenesis Alongside Chemoresistance. Asian Pac J Cancer Prev 2018;19(3):615–23 doi 10.22034/APJCP.2018.19.3.615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Larionova I, Kazakova E, Gerashchenko T, Kzhyshkowska J. New Angiogenic Regulators Produced by TAMs: Perspective for Targeting Tumor Angiogenesis. Cancers (Basel) 2021;13(13) doi 10.3390/cancers13133253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Zhao H, Chen Q, Alam A, Cui J, Suen KC, Soo AP, et al. The role of osteopontin in the progression of solid organ tumour. Cell Death Dis 2018;9(3):356 doi 10.1038/s41419-018-0391-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Tan Y, Zhao L, Yang YG, Liu W. The Role of Osteopontin in Tumor Progression Through Tumor-Associated Macrophages. Front Oncol 2022;12:953283 doi 10.3389/fonc.2022.953283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.McKay RR, Zurita AJ, Werner L, Bruce JY, Carducci MA, Stein MN, et al. A Randomized Phase II Trial of Short-Course Androgen Deprivation Therapy With or Without Bevacizumab for Patients With Recurrent Prostate Cancer After Definitive Local Therapy. J Clin Oncol 2016;34(16):1913–20 doi 10.1200/JCO.2015.65.3154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Humphrey PA, Halabi S, Picus J, Sanford B, Vogelzang NJ, Small EJ, et al. Prognostic significance of plasma scatter factor/hepatocyte growth factor levels in patients with metastatic hormone- refractory prostate cancer: results from cancer and leukemia group B 150005/9480. Clin Genitourin Cancer 2006;4(4):269–74 doi 10.3816/CGC.2006.n.006. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The clinical data are available from the NCI data Archive. The laboratory data are available from the first author.