Comparative Effectiveness of Immune Checkpoint Inhibitors vs Chemotherapy by Tumor Mutational Burden in Metastatic Castration-Resistant Prostate Cancer

Ryon P Graf; Virginia Fisher; Janick Weberpals; Ole Gjoerup; Marni B Tierno; Richard S P Huang; Nicolas Sayegh; Douglas I Lin; Kira Raskina; Alexa B Schrock; Eric Severson; James F Haberberger; Jeffrey S Ross; James Creeden; Mia A Levy; Brian M Alexander; Geoffrey R Oxnard; Neeraj Agarwal

doi:10.1001/jamanetworkopen.2022.5394

. 2022 Mar 31;5(3):e225394. doi: 10.1001/jamanetworkopen.2022.5394

Comparative Effectiveness of Immune Checkpoint Inhibitors vs Chemotherapy by Tumor Mutational Burden in Metastatic Castration-Resistant Prostate Cancer

Ryon P Graf ¹, Virginia Fisher ¹, Janick Weberpals ², Ole Gjoerup ¹, Marni B Tierno ¹, Richard S P Huang ¹, Nicolas Sayegh ³, Douglas I Lin ¹, Kira Raskina ¹, Alexa B Schrock ¹, Eric Severson ¹, James F Haberberger ¹, Jeffrey S Ross ^1,⁴, James Creeden ¹, Mia A Levy ^1,⁵, Brian M Alexander ¹, Geoffrey R Oxnard ¹, Neeraj Agarwal ^3,^✉

¹Foundation Medicine, Cambridge, Massachusetts

²Real World Data Collaborations, Personalized Healthcare Data, Analytics and Imaging, F. Hoffmann-La Roche, Basel, Switzerland

³Huntsman Cancer Institute, University of Utah, Salt Lake City

⁴Upstate Medical University, Syracuse, New York

⁵Rush University Medical Center, Chicago, Illinois

Accepted for Publication: February 02, 2022.

Published: March 31, 2022. doi:10.1001/jamanetworkopen.2022.5394

^✉

Corresponding Author: Neeraj Agarwal, MD, Huntsman Cancer Institute, University of Utah, 2000 Circle of Hope, Ste 2123, Salt Lake City, UT 84112 (neeraj.agarwal@hci.utah.edu).

Author Contributions: Dr Graf had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Graf, Gjoerup, Tierno, Raskina, Severson, Haberberger, Ross, Creeden, Levy, Agarwal.

Acquisition, analysis, or interpretation of data: Graf, Fisher, Weberpals, Tierno, Huang, Sayegh, Lin, Raskina, Schrock, Severson, Haberberger, Ross, Alexander, Oxnard, Agarwal.

Drafting of the manuscript: Graf, Fisher, Sayegh, Severson, Ross, Levy, Agarwal.

Critical revision of the manuscript for important intellectual content: Graf, Fisher, Weberpals, Gjoerup, Tierno, Huang, Sayegh, Lin, Raskina, Schrock, Severson, Haberberger, Ross, Creeden, Alexander, Oxnard, Agarwal.

Statistical analysis: Graf, Fisher, Weberpals, Raskina, Severson, Haberberger.

Administrative, technical, or material support: Gjoerup, Huang, Severson, Alexander.

Supervision: Severson, Creeden, Levy, Alexander, Oxnard, Agarwal.

Conflict of Interest Disclosures: Drs Graf, Fisher, Weberpals, Gjoerup, Tierno, Huang, Lin, Raskina, Schrock, Severson, Ross, Creeden, Levy, Alexander, and Oxnard and Mr Haberberger reported owning stock in Roche during the conduct of the study. Dr Agarwal reported receiving personal fees from Astellas, AstraZeneca, Aveo, Bayer, Bristol Myers Squibb, Calithera, Clovis, Eisai, Eli Lilly, EMD Serono, Exelixis, Foundation Medicine, Genentech, Gilead, Janssen, Merck, MEI Pharma, Nektar, Novartis, Pfizer, Pharmacyclics, and Seattle Genetics outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by Foundation Medicine. Foundation Medicine, a wholly owned subsidiary of Roche, is a for-profit company and producer of molecular diagnostics regulated by the US Food and Drug Administration. Roche produces the immune checkpoint inhibitor atezolizumab.

Role of the Funder/Sponsor: The authors are employed by Foundation Medicine and had roles in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Roberto Nussenzveig, PhD (Huntsman Cancer Institute, University of Utah, Salt Lake City), provided rigorous dialogue and feedback. Dr Nussenzveig was not compensated for this contribution.

^✉

Corresponding author.

PMCID: PMC8972027 PMID: 35357449

Key Points

Question

What is the comparative effectiveness of single-agent immune checkpoint inhibitors (ICIs) vs taxane chemotherapy in populations of patients with metastatic castration-resistant prostate cancer (mCRPC) defined by levels of tumor mutational burden (TMB)?

Findings

In this comparative effectiveness study of 741 patients with mCRPC, patients with TMB of 10 mutations per megabase (mt/Mb) or greater had significantly longer time to next treatment and overall survival with ICIs vs taxanes.

Meaning

These findings suggest that in scenarios where taxane use is considered, ICIs are a viable alternate treatment option for patients with mCRPC and TMB of 10 mt/Mb or greater.

This comparative effectiveness study assesses outcomes among patients with metastatic castration-resistant prostate cancer receiving immune checkpoint inhibitors (ICIs) vs taxane chemotherapy, stratified by tumor mutational burden (TMB).

Abstract

Importance

The most useful biomarkers for clinical decision-making identify patients likely to have improved outcomes with one treatment vs another.

Objective

To evaluate treatment class-specific outcomes of patients receiving immune checkpoint inhibitor (ICI) vs taxane chemotherapy by tumor mutational burden (TMB).

Design, Setting, and Participants

This comparative effectiveness analysis of clinical variables and outcomes used prospectively defined biomarker-stratified genomic data from a deidentified clinicogenomic database. Data included men with previously treated metastatic castration-resistant prostate cancer (mCRPC) receiving ICI or single-agent taxane chemotherapy from January 2011 to April 2021 at approximately 280 US academic or community-based cancer clinics (approximately 800 sites of care). Data were analyzed from July to August 2021.

Exposures

Single-agent ICI or single-agent taxanes. Treatments were assigned at discretion of physician and patient without randomization. Imbalances of known factors between treatment groups were adjusted with propensity weighting.

Main Outcomes and Measures

Prostate-specific antigen (PSA) response, time to next therapy (TTNT), and overall survival (OS).

Results

A total of 741 men (median [IQR], 70 [64-76] years) with mCRPC received comprehensive genomic profiling and were treated with ICI or single-agent taxane therapy. At baseline, the median (IQR) PSA level was 79.4 (19.0-254) ng/mL, 108 men (18.8%) had Eastern Cooperative Oncology Group Performance Status scores of 2 or greater, and 644 men (86.9%) had received prior systemic treatments for mCRPC. A total of 45 patients (6.1%) received ICI therapy and 696 patients (93.9%) received taxane therapy. Among patients with TMB of fewer than 10 mutations per megabase (mt/Mb) receiving ICI, compared with those receiving taxanes, had worse TTNT (median [IQR], 2.4 [1.1-3.2] months vs 4.1 [2.2-6.3] months; hazard ratio [HR], 2.65; 95% CI, 1.78-3.95; P < .001). In contrast, for patients with TMB of 10 mt/Mb or greater, use of ICIs, compared with use taxanes, was associated with more favorable TTNT (median [IQR], 8.0 [3.4 to unknown] months vs 2.4 [2.4-7.3] months; HR, 0.37, 95% CI, 0.15-0.87; P = .02) and OS (median 19.9 [8.06 to unknown] months vs 4.2 [2.69 – 6.12] months; HR, 0.23; 95% CI, 0.10-0.57; P = .001). Among all 741 patients, 44 (5.9%) had TMB of 10 mt/Mb or greater, 22 (3.0%) had high microsatellite instability, and 20 (2.7%) had both. Treatment interactions with TMB of 10 mt/Mb or greater (TTNT: HR, 0.10; 95% CI, 0.32-0.31; P < .001; OS: HR, 0.25; 95% CI, 0.076-0.81; P = .02) were stronger than high microsatellite instability alone (TTNT: HR, 0.12; 95% CI, 0.03-0.51; P = .004; OS: HR, 0.38; 95% CI, 0.13-1.12; P = .08).

Conclusions and Relevance

In this comparative effectiveness study, ICIs were more effective than taxanes in patients with mCRPC when TMB was 10 mt/Mb or greater but not when TMB was fewer than 10 mt/Mb. The results add validity to the existing TMB cutoff of 10 mt/Mb for ICI use in later lines of therapy, and suggest that ICIs may be a viable alternative to taxane chemotherapy for patients with mCRPC with high TMB.

Introduction

Immune checkpoint inhibitors (ICI) can enable deep responses and durable benefit in some patients with metastatic cancer who have received many prior treatments. However, the rate of clinical benefit differs considerably by tumor type. Unfortunately, for patients with metastatic castration-resistant prostate cancer (mCRPC) the objective response rate for ICI treatments has been reported as 3% for patients with tumors without programmed cell death ligand 1 (PD-L1) expression and 5% for those with PD-L1–expressing tumors.¹ For this reason, interest has developed in other biomarkers that might identify patients with mCRPC likely to receive greater clinical benefit from ICIs than alternate treatments. Pembrolizumab, an anti–programmed cell death 1 (PD1) drug, has received pantumor US Food and Drug Administration (FDA) approvals for patients with clinically advanced solid tumors who have progressed after prior treatment, have no satisfactory alternative treatment options,² and have tumors with high microsatellite instability (MSI-H) (ie, mismatch repair deficient) and patients with high tumor mutational burden (TMB) of at least 10 mutations per megabase (mt/Mb). These approvals were significantly influenced by the KEYNOTE-016 and KEYNOTE-158 studies.^3,4 KEYNOTE-016 enrolled only 2 patients with prostate cancer with MSI-H, observing stable disease and a partial response.³ KEYNOTE-158 observed objective responses in 30 of 102 patients in the group with TMB of 10 mt/Mb or greater and 43 patients of 688 patients in the group with TMB of fewer than 10 mt/Mb.⁴ While this included a wide range of tumor types, there were no patients with prostate cancer in the TMB of 10 mt/Mb or greater group.

To our knowledge, the largest single study to date evaluating outcomes among patients with MSI-H or mismatch repair deficient mCRPC receiving anti–PD-1 or anti–PD-L1 therapy reported declines of prostate-specific antigen (PSA) levels of 50% or more in 6 of 11 patients in a retrospective single-center study.⁵ Evaluating outcomes on a combination of ipilimumab plus nivolumab, a recent evaluation of the CheckMate-650 study reported an objective response rate of 19 of 33 patients (58%) with TMB of 10 mt/Mb or greater.⁶ Recently published supplemental data submitted to the FDA outside of KEYNOTE-158 supporting the pantumor TMB label (using a whole exome sequencing assay with high TMB calibrated to the existing companion diagnostic assay) reported prostate cancer objective responses in 1 of 11 patients (9%) in the high TMB and 7 of 115 patients (6%) in the low TMB group.² Notably, patients with mCRPC frequently have diffuse bone-predominant metastatic deposits, making many patients ineligible for objective response assessments.¹ Owing to this caveat, the KEYNOTE-199 trial included a cohort of patients with bone-predominant disease, and a trend for superior PFS and OS was observed in this cohort compared with those assessable for both response evaluation criteria in solid tumors and PD-L1.¹

While response rates are valuable surrogate outcomes for assessing efficacy, treatment decisions are always made in the context of an alternate path, such as drug A vs drug B. In the case of mCRPC, the alternative to ICI for most contexts of use is taxane chemotherapy (often as a rechallenge with limited clinical benefit) or palliative care. To our knowledge, there are no studies, prospective or retrospective, comparing outcomes of ICI vs an alternate drug class in mCRPC using established statistical criteria⁷ or evaluating overall survival (OS). In this study, we sought to compare clinical outcomes of patients receiving ICIs vs taxane chemotherapy in cohorts of patients with mCRPC with comparable characteristics and prior treatments and adjusted for known imbalances in treatment assignment and prognostic factors.

Methods

This comparative effectiveness study was approved by the Western Institutional Review Board, including a waiver of informed consent and a Health Insurance Portability and Accountability Act waiver of authorization. This study followed the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) reporting guideline.

Study Design and Patient Selection

The study cohort included patients with confirmed diagnosis of mCRPC included in the US-wide Flatiron Health and Foundation Medicine deidentified clinicogenomic database between January 2011 and April 2021 who underwent genomic testing using Foundation Medicine comprehensive genomic profiling (CGP) assays.

The deidentified data originated from approximately 280 US cancer clinics (approximately 800 sites of care). Retrospective longitudinal clinical data were derived from electronic health record data, comprising patient-level structured and unstructured data, curated via technology-enabled abstraction of clinical notes and radiology or pathology reports, and were linked to CGP data by deidentified, deterministic matching.⁸ Clinical data included demographics, clinical features, therapy exposure (with start and stop dates for each therapy line), and survival.

The clinicogenomic database contains multiple lines of therapy (LOT) per patient. Patients were included in this study if they received a single-agent anti–PD-1 axis therapy or single-agent taxane in the mCRPC setting and had TMB assessed via tissue biopsy. Some patients received a taxane and ICI on separate LOT; for these, the ICI LOT was chosen owing to less common ICI use. Some patients had more than 1 line of taxane given; for these patients, the latter-most taxane LOT was chosen to help mimic the FDA label for pembrolizumab² and restrict to patient populations that may be more homogeneous in their baseline characteristics, reducing potential bias in estimates.⁹ The patient selection flowchart is presented in Figure 1.

Figure 1. — CGP indicates comprehensive genomic profiling; ECOG PS, Eastern Cooperative Oncology Group Performance Score; ICI, immune checkpoint inhibitor; LOT, lines of therapy; mCRPC, metastatic castration-resistant prostate cancer; NHT, novel hormonal therapy; PD-1, programmed cell death 1; PSA, prostate-specific antigen; TMB, tumor mutational burden; and TTNT, time to next treatment.

Comprehensive Genomic Profiling

The FoundationOne or FoundationOne CDx hybrid capture-based next-generation sequencing assays were performed on patient tumor biopsies in a Clinical Laboratory Improvement Amendments–certified, College of American Pathologists–accredited laboratory (Foundation Medicine). Samples were evaluated for alterations as previously described.¹⁰ TMB was determined on up to 1.1 Mb of sequenced DNA.¹¹ MSI was determined on 95 to 114 loci, as previously described.¹²

PSA Response Assessment

A LOT was eligible for PSA response assessment if a PSA result was available within 60 days prior to LOT, and a separate PSA result was available 1 to 180 days after. If multiple results were available, respective values most proximal to treatment initiation, and 12 weeks of receiving treatment were used. PSA response was calculated as (receiving treatment PSA − baseline PSA) / (baseline PSA + 0.01)

Statistical Analysis

An a priori statistical analysis plan was developed and executed. The inclusion criteria, exclusion criteria, potential biases, primary outcome measures, exploratory outcome measures, handling of missing data, and all statistical methods were specified prior to analysis execution. The prespecified analysis compared the effectiveness of ICI vs taxane chemotherapy in patients stratified by TMB of fewer than 10 mt/Mb and TMB of 10 mt/Mb or greater by evaluating PSA response, time to next therapy (TTNT), and OS in concert. Stratification by MSI status was included as an exploratory analysis. Stratification by additional biomarkers was not prespecified. The statistical analysis plan also outlined a case-crossover analysis that was deemed not feasible owing to small sample size.

We used χ² tests and Wilcoxon rank sum tests to assess differences between cohorts of categorical and continuous variables, respectively. Missing values were imputed with the expected values based on observed covariates using random forests with R statistical software version 3.6.3 (R Project for Statistical Computing) package missForest, with imputed values treated identically to measured values in subsequent analysis. TTNT was calculated from start of treatment to start of next treatment, or last clinic visit if they had not yet reached next treatment. Patients were right-censored if they had not yet reached next treatment or died. OS was calculated from start of treatment to death from any cause, and patients with no record of mortality were right censored at the date of last clinic visit. Electronic health record OS data derivation has been previously described.¹³ OS risk intervals were instead initiated at time of CGP report (left truncation) if the CGP report was received after treatment initiation, as patients who die before a CGP report is delivered are not included in the database.¹⁴ Differences in time-to-event outcomes were assessed with the log-rank test and Cox proportional hazard models. Multiple comparison adjustments were not performed. P values are reported to quantify the strength of association for biomarker and each respective outcome, not for null hypothesis significance testing. Two-sided tests were used throughout. Results were interpreted considering all outcome measures collectively (PSA response, TTNT, and OS) rather than outcome measures standing on their own.

Propensity analyses are frequently used in observational data–heavy fields, such as epidemiology, economics, and the social sciences, as a best practice to adjust for subgroup imbalances.¹⁵ Propensity scores made use of the full matching technique (R package MatchIt) without caliper restrictions, resulting in no patient exclusions but taxane treatments receiving weights. Among patients receiving taxanes, those with characteristics most similar to the patients receiving ICIs were weighted more, and those less like the patients receiving ICIs weighted less. These weights were included in all Kaplan-Meier visualizations and Cox proportional hazard models. The features included for adjustment in the propensity model were age at time of treatment initiation, pretherapy PSA level, hemoglobin, alkaline phosphatase, Eastern Cooperative Oncology Group Performance Status (ECOG PS) score, TMB, mCRPC treatment line, prior novel hormonal therapy (NHT; yes vs no), and prior taxane (yes vs no). Full matching made use of a generalized linear model, and weights were truncated at 10 equivalents to limit influence per observation. Standardized mean difference (SMD) was used to assess balance, and within 10% was considered acceptable.¹⁵

Owing to focus on the 10 mt/Mb TMB threshold, propensity weights were created separately for the group with TMB of 10 mt/Mb or greater and the group with TMB of fewer than 10 mt/Mb for best possible within-group balance. For analyses evaluating MSI-H vs microsatellite stable (MSS), propensity weights were similarly created separately per group.

Biomarker treatment interaction tests⁷ made use of propensity-weighted multivariable Cox proportional hazards regression models containing the following variables: drug class (ICI or taxane), biomarker (high vs low), and the interaction term between drug class and biomarker. Hazard ratios (HRs) were then generated from adjusted Cox models stratified by group (ie, TMB high vs low). R version 3.6.3 software was used for all statistical analyses. Analyses were conducted from July to August 2021.

Results

After LOT selection, the cohort include 741 treatment exposures from 741 unique patients (median [IQR] age, 70 [64-76] years) (Figure 1 and Table), with baseline median (IQR) pretreatment PSA levels of 79.4 (19.0-254) ng/mL (to convert to micrograms per liter, multiply by 1.0). A total of 108 patients (18.8%) had tumors with ECOG PS scores of 2 or greater and 644 patients (86.9%) had received prior systemic treatments for mCRPC. A total of 45 patients (6.1%) received ICIs, and 741 patients (93.9%) received taxanes. There were no significant differences between patients who received ICIs vs those who received taxanes in age, pretherapy PSA levels, ECOG PS score, practice setting (community vs academic), prior NHT use, prior prescribed opioid use, and biopsy site. However, patients receiving ICIs, compared with those receiving taxane, had higher TMB (median [IQR], 3.5 [1.7-15.0] mt/Mb vs 2.5 [1.3-3.8] mt/Mb; P < .001).

Table. Patient Characteristics by Line of Treatment Among Patients with Metastatic Castration-Resistant Prostate Cancer.

Characteristic	No. (%)			P value
Characteristic	ICI (n = 45)	Taxane (n = 696)	Total (N = 741)	P value
Age, median (IQR), y	72.0 (64.0-76.0)	70.0 (64.0-76.0)	70.0 (64.0-76.0)	.72
TMB, median (IQR), mt/Mb	3.5 (1.7-15.0)	2.5 (1.3-3.8)	2.5 (1.3-3.8)	<.001
ECOG PS^a
0	5 (13.9)	158 (29.4)	163 (28.4)	.06
1	20 (55.6)	283 (52.6)	303 (52.8)
≥2	11 (30.6)	97 (18.0)	108 (18.8)
PSA, median (IQR), ng/mL	63.9 (10.8-183.2)	80.8 (19.4-256.1)	79.4 (19.0-254.0)	.21
Hemoglobin, median (IQR), g/dL	10.8 (9.5-11.9)	11.2 (9.8-12.6)	11.2 (9.7-12.6)	.11
Alkaline phosphatase, median (IQR), IU/L	91.0 (77.0-145.0)	110.0 (77.0-186.0)	109.0 (77.0-184.0)	.20
Treatment line
First	5 (11.1)	92 (13.2)	97 (13.1)	.06
Second	5 (11.1)	167 (24.0)	172 (23.2)
Third	9 (20.0)	167 (24.0)	176 (23.8)
≥Fourth	26 (57.8)	270 (38.8)	296 (39.9)
Practice type
Academic	5 (11.1)	102 (14.7)	107 (14.4)	.51
Community	40 (88.9)	594 (85.3)	634 (85.6)	.51
Prior NHT	35 (77.8)	552 (79.3)	587 (79.2)	.81
Prior taxane	33 (73.3)	374 (53.7)	407 (54.9)	.01
Prior opioid use	25 (55.6)	402 (57.8)	427 (57.6)	.77
Treatment received
Atezolizumab	2 (4.4)	0	2 (0.3)	<.001
Cabazitaxel	0	342 (49.1)	342 (46.2)
Docetaxel	0	354 (50.9)	354 (47.8)
Nivolumab	9 (20.0)	0	9 (1.2)
Pembrolizumab	34 (75.6)	0	34 (4.6)
Biopsy site
Liver	5 (11.)	92 (13.2)	97 (13.1)	.11
Lymph node	2 (4.4)	120 (17.2)	122 (16.5)
Other	14 (31.1)	199 (28.6)	213 (28.7)
Prostate	24 (53.3)	285 (40.9)	309 (41.7)

Open in a new tab

Abbreviations: ECOG PS, Eastern Cooperative Oncology Group Performance Status; ICI, immune checkpoint inhibitor; mt/Mb, mutations per megabase; NHT, novel hormonal therapy; PSA, prostate-specific antigen; TMB, tumor mutational burden.

SI conversion factor: To convert alkaline phosphatase to microkatals per liter, multiply by 0.0167; hemoglobin to grams per liter, multiply by 10; prostate-specific antigen to micrograms per liter, multiply by 1.

^{^a}

Among 574 patients with data.

Propensity Weighting

Despite favorable overall cohort balance prior to propensity weighting, 6 of 9 features had greater than 10% SMD. After weighting, only TMB had greater than 10% SMD, with patients receiving ICI having higher values (eFigure 1 in the Supplement). The main source of this remaining imbalance was the subgroup with TMB of 10 mt/Mb or greater, as the subgroup with TMB of 10 mt/Mb or fewer had less of an imbalance (eFigure 1 in the Supplement). The subgroup with TMB of 10 mt/Mb or greater additionally had residual imbalance of higher ECOG PS scores for the ICI group. TMB was not significantly associated with TTNT or OS among patients receiving taxane chemotherapy (eFigure 2 in the Supplement).

PSA Response

PSA response was evaluable in 607 patients (81.9%) (Figure 1). The most common reason for unevaluable PSA response was lack of baseline PSA value (eTable 1 in the Supplement). For 14 patients with TMB of 10 mt/Mb or greater who received ICI, only 9 had evaluable PSA responses. Of these, 4 patients had PSA declines of greater than 50% from baseline. Of 31 patients with TMB of 10 mt/Mb or fewer who received an ICI, 24 were evaluable for PSA responses, and none had PSA declines of greater than 50% from baseline (Figure 2A). For patients receiving taxanes, TMB level was not associated with PSA response (Figure 2B).

Figure 2. — Waterfall plots indicate the relative change in PSA from baseline to approximately 12 weeks of therapy for patients receiving (A) single-agent anti–programmed cell death 1 (PD-1) axis therapies and (B) single-agent taxanes. A total of 134 patients did not have evaluable PSA response, and are not represented in these plots, including 5 patients with tumor mutational burden (TMB) of 10 mutations per megabase (mt/Mb) who received single-agent anti–PD-1 axis therapy. Patients with more than 100% PSA gain are capped at 100.

Outcomes by Therapy Class and TMB of 10 mt/Mb or Greater

The propensity-adjusted TTNT and OS by drug class was stratified by whether patients had TMB of fewer than 10 mt/Mb or 10 mt/MB or greater, the FDA-approved cutoff indicated on the pantumor label of pembrolizumab.² Patients with TMB of fewer than 10 mt/Mb receiving ICIs, compared with those receiving taxanes, had worse TTNT (median [IQR], 2.4 [1.1-3.2] months vs 4.1 [2.2-6.3] months; HR, 2.65, 95% CI, 1.78-3.85; P < .001), while the trend was reversed for patients with TMB of 10 mt/Mb or greater, with observed improved TTNT (median [IQR], 8.0 [3.4 to unknown] months vs 2.4 [2.43-7.33] months; HR, 0.37; 95% CI, 0.15-0.87; P = .02) (Figure 3). There was no significant difference between patients with TMB of fewer than 10 mt/Mb receiving ICIs vs those receiving taxanes in OS (median [IQR], 4.2 [2.1-8.1] months vs 6.0 [3.1-9.7] months; HR, 1.08; 95% CI, 0.68-1.74; P = .73), but for patients with TMB of 10 mt/Mb or greater, those receiving ICIs had better OS (median [IQR], 19.9 [8.1 to unknown] months vs 4.2 [2.69-6.12] months; HR, 0.23, 95% CI, 0.10-0.57; P = .001). Treatment interactions with TMB were observed for both TTNT (HR, 0.10; 95% CI, 0.03-0.31; P < .001) and OS (HR, 0.25; 95% CI, 0.08-0.81; P = .02). As a sensitivity analysis, TTNT and OS analyses were additionally conducted without propensity adjustments, with similar results observed (eFigure 3 in the Supplement). As an additional sensitivity analysis, the prognoses of patients with tumors having TMB of less than 10 mt/Mb or TMB of 10 mt/Mb or greater and receiving ICI or taxanes were compared (eFigure 4 in the Supplement), with TMB being highly prognostic of TTNT and OS for patients receiving ICI but not taxanes, unadjusted and adjusted for potential confounding factors, such as ECOG PS score, PSA level, and prior treatments (eTable 2 in the Supplement).

Figure 3. — Overall survival estimates are left truncated with at-risk tables adjusted accordingly. Numerical summary and treatment interaction tests can be found in eTable 3 in the Supplement. HR indicates hazard ratio; mt/Mb, mutations per megabase; PD-1, programmed cell death 1; TMB, tumor mutational burden.

TMB, MSI, and Treatment Interactions

The association between TMB and MSI was additionally explored. Of all 741 patients, 2 (<1%) had MSI-H and TMB of fewer than 10 mt/Mb, 20 (2.7%) had MSI-H and TMB of 10 mt/Mb or greater, 24 (3.2%) had MSS and TMB of 10 mt/Mb or greater, 601 (81%) had MSS and TMB of fewer than 10 mt/Mb, and 94 (12.7%) had TMB of fewer than 10 mt/Mb and had unknown or indeterminate microsatellite status (eTable 4 in the Supplement). Comparing outcomes of ICI vs taxanes in the MSI-H group, those receiving ICIs had more favorable TTNT (HR, 0.38; 95% CI, 0.15-0.94; P = .04) (Figure 4A), but there was no significant difference in OS (HR, 0.44; 95% CI, 0.15-1.27; P = .13) (Figure 4B). Patients with TMB of 10 mt/Mb or greater receiving ICIs had better OS and TTNT (Figure 3B, 3D, Figure 4A, 4B; eFigure 5 and eFigure 6 in the Supplement) with weaker treatment interactions (Figure 4A, 4B).

However, only 15 total patients had TMB of 10 mt/Mb or greater and/or MSI-H and received ICIs (Figure 4C; eTable 5 in the Supplement). The only patient with MSI-H and TMB of fewer than 10 mt/Mb who received an ICI had a PSA increase of 210%, TTNT of 4.8 months, and OS of 5.1 months. Of 3 patients in the TMB of 10 mt/Mb or greater but MSS subgroup, 1 had a PSA increase of 20% and was still receiving an ICI after 1.3 months, 1 was unevaluable for PSA response and had TTNT of 4.1 months and OS of 8.1 months, and 1 had a PSA reduction of 78% and had not initiated a new treatment nor died at 13.1 months after treatment initiation at time of data lock.

While the etiology of TMB levels was not a focus of this study, among 24 patients who had TMB of 10 mt/Mb or greater without MSI-H, 10 had a deleterious alteration in a known MMR gene and 14 did not (eTable 6 in the Supplement). While the FDA label for TMB is not dependent upon the etiology of TMB levels, we included subgroup analyses of patients with MSS and TMB of 10 mt/Mb or greater with respect to TTNT and OS (eTable 7 in the Supplement). Because there were 3 patients in this group receiving ICIs, and only 1 patient had a TTNT or OS event, the point estimates for TTNT (HR, 0.44; 95% CI, 0.04-4.67) and OS (HR, 0.50; 95% CI, 0.10-2.65), while numerically consistent with the broader TMB of 10 mt/Mb or greater subgroup, must be interpreted in the context of these caveats.

Discussion

This comparative effectiveness study evaluating a cohort of patients treated with ICIs or taxanes found improved TTNT and OS in patients with high TMB treated with ICIs vs taxanes. Treatment decisions are always dependent on multiple courses of action. The most useful biomarkers are those that can identify patients who are anticipated to have differing outcomes on different viable treatment paths, such as 2 different FDA-approved drugs.

In practice, the decision to use ICIs vs taxanes will also weigh nonbiomarker considerations, such as patient frailty, cost, and tolerability. While there exists no head-to-head comparison of adverse events (AEs) of ICIs vs taxanes, both the KEYNOTE-199 and CARD studies were conducted in the post-NHT, postdocetaxel mCRPC setting. KEYNOTE-199 reported AEs for pembrolizumab in 60% of patients, with 15% having grade 3 to 5 AEs.¹ The cabazitaxel group of the CARD study reported AEs in 98% of patients, with 56% of patients experiencing grade 3 to 5 AEs.¹⁶

When a treatment is working well for a patient, their TTNT will be prolonged. However, a potential source of bias for TTNT assessments is when 1 regimen has more AEs than the other; patients with very poor benefit might need additional time to recover after their treatment discontinuation from a less tolerable regimen. For these reasons, TTNT values for taxanes relative to ICIs taken at face value might result in an inflated evaluation of taxane benefit relative to ICIs.

Within the subgroup of patients with TMB of 10 mt/Mb or greater, there were residual imbalances after weighting indicating higher TMB and higher ECOG PS scores for the patients receiving ICIs. While higher TMB was not associated with poorer survival in patients receiving taxanes in this study or across non-ICI treatments in many cancer types,¹⁷ poor performance status is a negative prognostic marker in mCRPC.¹⁸ The residual known imbalances in the subgroup with TMB of 10 mt/Mb or greater would potentially result in a relative penalty toward ICI outcomes relative to taxanes in this subgroup, even after adjustment.

Patients in clinical settings often differ from those enrolled in the randomized clinical trials.¹⁹ We evaluated the outcomes of ICIs vs taxanes in a heavily pretreated cohort, in which 79.2% of patients had received prior NHT and 54.9% of patients had received prior taxanes. Among those treated with taxanes, only few had PSA declines of 50% or greater, and less than half had any PSA decline. This context of use calls into question the true utility of systemic cytotoxic therapy in this setting. Comparisons of ICIs vs taxanes in this setting could be influenced by clinical practices particular to this context for use, and care must be taken for extrapolation of our results outside of this context. However, given this context, and carefully considering biases and imbalances, our results are consistent with the hypothesis that ICIs are a superior option to taxanes for patients with TMB of 10 mt/Mb or greater. Whether an ICI might be a superior option to other treatment modalities, such as second generation NHTs, for patients with TMB of 10 mt/Mb or greater in earlier LOT remains unevaluated. In our cohort, most patients with TMB of 10 mt/Mb or greater who received ICIs also had MSI-H. Among 3 patients with TMB of 10 mt/Mb or greater MSS in this group, 2 had greater than median TTNT for patients with TMB of 10 mt/Mb or greater receiving taxanes, and the third had only 1.3 months of follow-up without a treatment switch, making interpretations challenging.

While to our knowledge, our study is the most rigorous analysis to date comparing ICI and taxane effectiveness in biomarker groups consistent with FDA labels, prospective randomized evaluation would add precision and granularity to comparative efficacy estimates for ICIs vs taxanes for patients with high TMB, potentially stratified by MSI status, MMR gene alterations, and in earlier lines of therapy.

The evaluation of TMB is dependent on the sequencing breadth, depth, and characteristics of bioinformatic germline filtering, on top of the degree of platform analytical validity and demonstrated rigor and robustness of measurement.²⁰ For these reasons, the clinical validity (associations with patient outcomes, such as in the American Society of Clinical Oncology and College of American Pathologists guideline²¹) cannot be assumed to transfer from platform to platform. It is very important that the biomarker platform evaluated for outcome associations be the one used in practice to guide therapy. A strength of the study is the use of the platform supporting the only existing FDA-approved TMB companion diagnostic.

Rigorous genomics-centered comparative effectiveness studies of standard of care patient populations are likely to become increasingly important in the coming years, as the systemic treatment landscape for many cancer types will continue to diversify. Considering mCRPC as an example, recent FDA approvals have resulted in Poly (ADP-ribose) polymerase inhibitors and ICIs being added alongside the existing standard of care therapies of NHT and taxane chemotherapy. In the near future, anti–prostate-specific membrane antigen radionuclides²² are likely to be approved, while other treatment modalities, such as AKT inhibitors²³ and androgen receptor degraders,²⁴ show promise. It will not be possible for control groups of future registrational trials to rigorously assess all alternatives in a diversified landscape. The use of clinical data, especially linked to biomarkers, can help fill knowledge gaps to aid in decision-making in a wide variety of clinical scenarios.

Limitations

This study has some limitations. This is not a randomized clinical trial. Treatment assignments were at the discretion of the clinician, and while biases were carefully considered and known imbalances adjusted, unknown imbalances likely remain. The number of patients with mCRPC receiving ICI was relatively small. Patients were not restricted by timing of biopsy (archival vs contemporaneous) and higher TMB is more common in contemporaneous biopsies. Additionally, 40.9% of the cohort specimens were from prostate anatomical origin, and it is possible that there may exist false negatives with respect to TMB threshold assessments.

Conclusions

The findings of this comparative effectiveness study add validity to the existing FDA-approved TMB cutoff of 10 mt/Mb. These results suggest that ICI may be a viable alternative to taxane chemotherapy in later lines of therapy for patients with mCRPC with high TMB.

Supplement.

eFigure 1. Propensity Weighting and Balance Adjustment

eTable 1. Sensitivity Analysis: PSA Response Evaluable vs Unevaluable

eFigure 2. TTNT and OS by Therapy Class and MSI-H Status

eFigure 3. Unadjusted TTNT and OS by Therapy Class

eFigure 4. Prognosis of Patients Receiving ICIs or Taxanes by TMB

eTable 2. Multivariable Assessment of Prognosis of Patients Receiving ICIs and Taxanes

eFigure 5. TTNT and OS by Therapy Class and MSI-H Status

eFigure 6. Propensity Weighting and Balance Adjustment

eTable 3. Stratified Cox PH Models and Cox PH Interaction Models

eTable 4. Associations of TMB, MSI, and Treatment Assigned

eTable 5. Patient Characteristics for TMB of 10 mt/Mb or Greater and/or MSI-H

eTable 6. Association of MMR Gene Deleterious Alterations With TMB of 10 mt/Mb or Greater and Non-MSI-H

eTable 7. MSS Only Subgroup Analyses

Click here for additional data file.^{(2.1MB, pdf)}

References

1.Antonarakis ES, Piulats JM, Gross-Goupil M, 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):395-405. doi: 10.1200/JCO.19.01638 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Marcus L, Fashoyin-Aje LA, Donoghue M, et al. FDA approval summary: pembrolizumab for the treatment of tumor mutational burden-high solid tumors. Clin Cancer Res. 2021;27(17):4685-4689. doi: 10.1158/1078-0432.CCR-21-0327 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Lemery S, Keegan P, Pazdur R. First FDA approval agnostic of cancer site—when a biomarker defines the indication. N Engl J Med. 2017;377(15):1409-1412. doi: 10.1056/NEJMp1709968 [DOI] [PubMed] [Google Scholar]
4.Marabelle A, Fakih M, Lopez J, et al. Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 study. Lancet Oncol. 2020;21(10):1353-1365. doi: 10.1016/S1470-2045(20)30445-9 [DOI] [PubMed] [Google Scholar]
5.Abida W, Cheng ML, Armenia J, et al. Analysis of the prevalence of microsatellite instability in prostate cancer and response to immune checkpoint blockade. JAMA Oncol. 2019;5(4):471-478. doi: 10.1001/jamaoncol.2018.5801 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Sharma P, Pachynski RK, Narayan V, et al. Nivolumab plus ipilimumab for metastatic castration-resistant prostate cancer: preliminary analysis of patients in the CheckMate 650 Trial. Cancer Cell. 2020;38(4):489-499.e3. doi: 10.1016/j.ccell.2020.08.007 [DOI] [PubMed] [Google Scholar]
7.Ballman KV. Biomarker: predictive or prognostic? J Clin Oncol. 2015;33(33):3968-3971. doi: 10.1200/JCO.2015.63.3651 [DOI] [PubMed] [Google Scholar]
8.Singal G, Miller PG, Agarwala V, et al. Association of patient characteristics and tumor genomics with clinical outcomes among patients with non-small cell lung cancer using a clinicogenomic database. JAMA. 2019;321(14):1391-1399. doi: 10.1001/jama.2019.3241 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Schneeweiss S, Patrick AR, Stürmer T, et al. Increasing levels of restriction in pharmacoepidemiologic database studies of elderly and comparison with randomized trial results. Med Care. 2007;45(10)(suppl 2):S131-S142. doi: 10.1097/MLR.0b013e318070c08e [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Frampton GM, Fichtenholtz A, Otto GA, et al. Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nat Biotechnol. 2013;31(11):1023-1031. doi: 10.1038/nbt.2696 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Chalmers ZR, Connelly CF, Fabrizio D, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017;9(1):34. doi: 10.1186/s13073-017-0424-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Trabucco SE, Gowen K, Maund SL, et al. A novel next-generation sequencing approach to detecting microsatellite instability and pan-tumor characterization of 1000 microsatellite instability-high cases in 67,000 patient samples. J Mol Diagn. 2019;21(6):1053-1066. doi: 10.1016/j.jmoldx.2019.06.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Curtis MD, Griffith SD, Tucker M, et al. Development and validation of a high-quality composite real-world mortality endpoint. Health Serv Res. 2018;53(6):4460-4476. doi: 10.1111/1475-6773.12872 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.McGough SF, Incerti D, Lyalina S, Copping R, Narasimhan B, Tibshirani R. Penalized regression for left-truncated and right-censored survival data. Stat Med. 2021;40(25):5487-5500. doi: 10.1002/sim.9136 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med. 2015;34(28):3661-3679. doi: 10.1002/sim.6607 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.de Wit R, de Bono J, Sternberg CN, et al. ; CARD Investigators . Cabazitaxel versus abiraterone or enzalutamide in metastatic prostate cancer. N Engl J Med. 2019;381(26):2506-2518. doi: 10.1056/NEJMoa1911206 [DOI] [PubMed] [Google Scholar]
17.Shao C, Li G, Huang L, et al. Prevalence of high tumor mutational burden and association with survival in patients with less common solid tumors. JAMA Netw Open. 2020;3(10):e2025109. doi: 10.1001/jamanetworkopen.2020.25109 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Halabi S, Lin CY, Kelly WK, 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):671-677. doi: 10.1200/JCO.2013.52.3696 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Yoo SH, Keam B, Kim M, Kim TM, Kim DW, Heo DS. Generalization and representativeness of phase III immune checkpoint blockade trials in non-small cell lung cancer. Thorac Cancer. 2018;9(6):736-744. doi: 10.1111/1759-7714.12641 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Vega DM, Yee LM, McShane LM, et al. ; TMB Consortium . Aligning tumor mutational burden (TMB) quantification across diagnostic platforms: phase II of the Friends of Cancer Research TMB Harmonization Project. Ann Oncol. 2021;32(12):1626-1636. doi: 10.1016/j.annonc.2021.09.016 [DOI] [PubMed] [Google Scholar]
21.Merker JD, Oxnard GR, Compton C, et al. Circulating tumor DNA analysis in patients with cancer: American Society of Clinical Oncology and College of American Pathologists Joint Review. J Clin Oncol. 2018;36(16):1631-1641. doi: 10.1200/JCO.2017.76.8671 [DOI] [PubMed] [Google Scholar]
22.Sartor O, de Bono J, Chi KN, et al. ; VISION Investigators . Lutetium-177-PSMA-617 for metastatic castration-resistant prostate cancer. N Engl J Med. 2021;385(12):1091-1103. doi: 10.1056/NEJMoa2107322 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Sweeney C, Bracarda S, Sternberg CN, et al. Ipatasertib plus abiraterone and prednisolone in metastatic castration-resistant prostate cancer (IPATential150): a multicentre, randomised, double-blind, phase 3 trial. Lancet. 2021;398(10295):131-142. doi: 10.1016/S0140-6736(21)00580-8 [DOI] [PubMed] [Google Scholar]
24.Kregel S, Wang C, Han X, et al. Androgen receptor degraders overcome common resistance mechanisms developed during prostate cancer treatment. Neoplasia. 2020;22(2):111-119. doi: 10.1016/j.neo.2019.12.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eFigure 1. Propensity Weighting and Balance Adjustment

eTable 1. Sensitivity Analysis: PSA Response Evaluable vs Unevaluable

eFigure 2. TTNT and OS by Therapy Class and MSI-H Status

eFigure 3. Unadjusted TTNT and OS by Therapy Class

eFigure 4. Prognosis of Patients Receiving ICIs or Taxanes by TMB

eTable 2. Multivariable Assessment of Prognosis of Patients Receiving ICIs and Taxanes

eFigure 5. TTNT and OS by Therapy Class and MSI-H Status

eFigure 6. Propensity Weighting and Balance Adjustment

eTable 3. Stratified Cox PH Models and Cox PH Interaction Models

eTable 4. Associations of TMB, MSI, and Treatment Assigned

eTable 5. Patient Characteristics for TMB of 10 mt/Mb or Greater and/or MSI-H

eTable 6. Association of MMR Gene Deleterious Alterations With TMB of 10 mt/Mb or Greater and Non-MSI-H

eTable 7. MSS Only Subgroup Analyses

Click here for additional data file.^{(2.1MB, pdf)}

[zoi220179r1] 1.Antonarakis ES, Piulats JM, Gross-Goupil M, 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):395-405. doi: 10.1200/JCO.19.01638 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r2] 2.Marcus L, Fashoyin-Aje LA, Donoghue M, et al. FDA approval summary: pembrolizumab for the treatment of tumor mutational burden-high solid tumors. Clin Cancer Res. 2021;27(17):4685-4689. doi: 10.1158/1078-0432.CCR-21-0327 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r3] 3.Lemery S, Keegan P, Pazdur R. First FDA approval agnostic of cancer site—when a biomarker defines the indication. N Engl J Med. 2017;377(15):1409-1412. doi: 10.1056/NEJMp1709968 [DOI] [PubMed] [Google Scholar]

[zoi220179r4] 4.Marabelle A, Fakih M, Lopez J, et al. Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 study. Lancet Oncol. 2020;21(10):1353-1365. doi: 10.1016/S1470-2045(20)30445-9 [DOI] [PubMed] [Google Scholar]

[zoi220179r5] 5.Abida W, Cheng ML, Armenia J, et al. Analysis of the prevalence of microsatellite instability in prostate cancer and response to immune checkpoint blockade. JAMA Oncol. 2019;5(4):471-478. doi: 10.1001/jamaoncol.2018.5801 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r6] 6.Sharma P, Pachynski RK, Narayan V, et al. Nivolumab plus ipilimumab for metastatic castration-resistant prostate cancer: preliminary analysis of patients in the CheckMate 650 Trial. Cancer Cell. 2020;38(4):489-499.e3. doi: 10.1016/j.ccell.2020.08.007 [DOI] [PubMed] [Google Scholar]

[zoi220179r7] 7.Ballman KV. Biomarker: predictive or prognostic? J Clin Oncol. 2015;33(33):3968-3971. doi: 10.1200/JCO.2015.63.3651 [DOI] [PubMed] [Google Scholar]

[zoi220179r8] 8.Singal G, Miller PG, Agarwala V, et al. Association of patient characteristics and tumor genomics with clinical outcomes among patients with non-small cell lung cancer using a clinicogenomic database. JAMA. 2019;321(14):1391-1399. doi: 10.1001/jama.2019.3241 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r9] 9.Schneeweiss S, Patrick AR, Stürmer T, et al. Increasing levels of restriction in pharmacoepidemiologic database studies of elderly and comparison with randomized trial results. Med Care. 2007;45(10)(suppl 2):S131-S142. doi: 10.1097/MLR.0b013e318070c08e [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r10] 10.Frampton GM, Fichtenholtz A, Otto GA, et al. Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nat Biotechnol. 2013;31(11):1023-1031. doi: 10.1038/nbt.2696 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r11] 11.Chalmers ZR, Connelly CF, Fabrizio D, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017;9(1):34. doi: 10.1186/s13073-017-0424-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r12] 12.Trabucco SE, Gowen K, Maund SL, et al. A novel next-generation sequencing approach to detecting microsatellite instability and pan-tumor characterization of 1000 microsatellite instability-high cases in 67,000 patient samples. J Mol Diagn. 2019;21(6):1053-1066. doi: 10.1016/j.jmoldx.2019.06.011 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r13] 13.Curtis MD, Griffith SD, Tucker M, et al. Development and validation of a high-quality composite real-world mortality endpoint. Health Serv Res. 2018;53(6):4460-4476. doi: 10.1111/1475-6773.12872 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r14] 14.McGough SF, Incerti D, Lyalina S, Copping R, Narasimhan B, Tibshirani R. Penalized regression for left-truncated and right-censored survival data. Stat Med. 2021;40(25):5487-5500. doi: 10.1002/sim.9136 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r15] 15.Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med. 2015;34(28):3661-3679. doi: 10.1002/sim.6607 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r16] 16.de Wit R, de Bono J, Sternberg CN, et al. ; CARD Investigators . Cabazitaxel versus abiraterone or enzalutamide in metastatic prostate cancer. N Engl J Med. 2019;381(26):2506-2518. doi: 10.1056/NEJMoa1911206 [DOI] [PubMed] [Google Scholar]

[zoi220179r17] 17.Shao C, Li G, Huang L, et al. Prevalence of high tumor mutational burden and association with survival in patients with less common solid tumors. JAMA Netw Open. 2020;3(10):e2025109. doi: 10.1001/jamanetworkopen.2020.25109 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r18] 18.Halabi S, Lin CY, Kelly WK, 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):671-677. doi: 10.1200/JCO.2013.52.3696 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r19] 19.Yoo SH, Keam B, Kim M, Kim TM, Kim DW, Heo DS. Generalization and representativeness of phase III immune checkpoint blockade trials in non-small cell lung cancer. Thorac Cancer. 2018;9(6):736-744. doi: 10.1111/1759-7714.12641 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r20] 20.Vega DM, Yee LM, McShane LM, et al. ; TMB Consortium . Aligning tumor mutational burden (TMB) quantification across diagnostic platforms: phase II of the Friends of Cancer Research TMB Harmonization Project. Ann Oncol. 2021;32(12):1626-1636. doi: 10.1016/j.annonc.2021.09.016 [DOI] [PubMed] [Google Scholar]

[zoi220179r21] 21.Merker JD, Oxnard GR, Compton C, et al. Circulating tumor DNA analysis in patients with cancer: American Society of Clinical Oncology and College of American Pathologists Joint Review. J Clin Oncol. 2018;36(16):1631-1641. doi: 10.1200/JCO.2017.76.8671 [DOI] [PubMed] [Google Scholar]

[zoi220179r22] 22.Sartor O, de Bono J, Chi KN, et al. ; VISION Investigators . Lutetium-177-PSMA-617 for metastatic castration-resistant prostate cancer. N Engl J Med. 2021;385(12):1091-1103. doi: 10.1056/NEJMoa2107322 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi220179r23] 23.Sweeney C, Bracarda S, Sternberg CN, et al. Ipatasertib plus abiraterone and prednisolone in metastatic castration-resistant prostate cancer (IPATential150): a multicentre, randomised, double-blind, phase 3 trial. Lancet. 2021;398(10295):131-142. doi: 10.1016/S0140-6736(21)00580-8 [DOI] [PubMed] [Google Scholar]

[zoi220179r24] 24.Kregel S, Wang C, Han X, et al. Androgen receptor degraders overcome common resistance mechanisms developed during prostate cancer treatment. Neoplasia. 2020;22(2):111-119. doi: 10.1016/j.neo.2019.12.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Comparative Effectiveness of Immune Checkpoint Inhibitors vs Chemotherapy by Tumor Mutational Burden in Metastatic Castration-Resistant Prostate Cancer

Ryon P Graf, PhD

Virginia Fisher, PhD

Janick Weberpals, RPh, PhD

Ole Gjoerup, PhD

Marni B Tierno, PhD, RN

Richard S P Huang, MD

Nicolas Sayegh, MD

Douglas I Lin, MD, PhD

Kira Raskina, MD, MS

Alexa B Schrock, PhD

Eric Severson, MD, PhD

James F Haberberger, BS

Jeffrey S Ross, MD

James Creeden, MD, PhD

Mia A Levy, MD, PhD

Brian M Alexander, MD, MPH

Geoffrey R Oxnard, MD

Neeraj Agarwal, MD

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting, and Participants

Exposures

Main Outcomes and Measures

Results

Conclusions and Relevance

Introduction

Methods

Study Design and Patient Selection

Figure 1. Patient Recruitment Flowchart.

Comprehensive Genomic Profiling

PSA Response Assessment

Statistical Analysis

Results

Table. Patient Characteristics by Line of Treatment Among Patients with Metastatic Castration-Resistant Prostate Cancer.

Propensity Weighting

PSA Response

Figure 2. Prostate-Specific Antigen (PSA) Response by Drug Class.

Outcomes by Therapy Class and TMB of 10 mt/Mb or Greater

Figure 3. Time to Next Treatment (TTNT) and Overall Survival (OS) by Therapy Class.

TMB, MSI, and Treatment Interactions

Figure 4. Time to Next Treatment (TTNT) and Overall Survival (OS) by Therapy Class, Tumor Mutational Burden (TMB), and Microsatellite Instability (MSI) Status.

Discussion

Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases