Does biomarker use in oncology improve clinical trial failure risk? A large-scale analysis

Abstract

Purpose: To date there has not been an extensive analysis of the outcomes of biomarker use in oncology.

Methods: Data were pooled across four indications in oncology drawing upon trial outcomes from www.clinicaltrials.gov: breast cancer, non-small cell lung cancer (NSCLC), melanoma and colorectal cancer from 1998 to 2017. We compared the likelihood drugs would progress through the stages of clinical trial testing to approval based on biomarker status. This was done with multi-state Markov models, tools that describe the stochastic process in which subjects move among a finite number of states.

Results: Over 10000 trials were screened, which yielded 745 drugs. The inclusion of biomarker status as a covariate significantly improved the fit of the Markov model in describing the drug trajectories through clinical trial testing stages. Hazard ratios based on the Markov models revealed the likelihood of drug approval with biomarkers having nearly a fivefold increase for all indications combined. A 12, 8 and 7-fold hazard ratio was observed for breast cancer, melanoma and NSCLC, respectively. Markov models with exploratory biomarkers outperformed Markov models with no biomarkers.

Conclusion: This is the first systematic statistical evidence that biomarkers clearly increase clinical trial success rates in three different indications in oncology. Also, exploratory biomarkers, long before they are properly validated, appear to improve success rates in oncology. This supports early and aggressive adoption of biomarkers in oncology clinical trials.

Keywords: biomarkers; breast cancer; cancer; clinical trial; drug development; lung cancer; melanoma; oncology; risk.

FIGURE 1

A depiction of five states of clinical trial testing that was used in Markov modelling. Different states used in the Markov model. Biomarker status was a covariate in one model while the other Markov model had no such covariate.

FIGURE 2

Breast cancer. The ability of two different Markov models to predict clinical trial successes in historical data in this indication. 'Biomarker' shows the performance of a Markov model with biomarker status as a covariate while non‐biomarker has no such covariate. Hazard represents the likelihood and rate of advancing to the next phase. Bars are 95% CI. Sample sizes: Phase I (n = 183), Phase II (n = 132) and Phase III (n = 49).

FIGURE 3

Colorectal. The ability of two different Markov models to predict clinical trial successes in historical data in this indication. 'Biomarker' shows the performance of a Markov model with biomarker status as a covariate while non‐biomarker has no such covariate. Hazard represents the likelihood and rate of advancing to the next phase. Bars are 95% CI. Sample sizes: Phase I (n = 195), Phase II (n = 128) and Phase III (n = 24).

FIGURE 4

Melanoma. The ability of two different Markov models to predict clinical trial successes in historical data in this indication. 'Biomarker' shows the performance of a Markov model with biomarker status as a covariate while non‐biomarker has no such covariate. Hazard represents the likelihood and rate of advancing to the next phase. Bars are 95% CI. Sample sizes: Phase I (n = 81), Phase II (n = 49) and Phase III (n = 21).

FIGURE 5

NSCLC. The ability of two different Markov models to predict clinical trial successes in historical data in this indication. 'Biomarker' shows the performance of a Markov model with biomarker status as a covariate while non‐biomarker has no such covariate. Hazard represents the likelihood and rate of advancing to the next phase. Bars are 95% CI. Sample sizes: Phase I (n = 286), Phase II (n = 217) and Phase III (n = 65).

FIGURE 6

A sub‐analysis of biomarkers into exploratory and validated biomarkers, respectively. The performance of Markov models using each of these biomarkers as covariates is depicted alongside Markov models that do not look at biomarkers. Clinical studies with exploratory biomarkers outperform clinical trials with no biomarkers. Total study sample size of 748 drugs was segmented into: no biomarker (n = 555 drugs), validated biomarker (n = 80) and exploratory biomarker‐based drugs (n = 113). Bars are 95% CI.