Lessons learned from BATTLE-2 in the war on cancer: the use of Bayesian method in clinical trial design

A Bayesian-adaptive decision-theoretic approach can reduce the sample sizes for multiarm exercise oncology trials

OBJECTIVES

Adaptive designs may reduce trial sample sizes and costs. This study illustrates a Bayesian-adaptive decision-theoretic design applied to a multiarm exercise oncology trial.

STUDY DESIGN AND SETTING

In the Physical exercise during Adjuvant Chemotherapy Effectiveness Study (PACES) trial, 230 breast cancer patients receiving chemotherapy were randomized to supervised resistance and aerobic exercise (OnTrack), home-based physical activity (OncoMove) or usual care (UC). Data were reanalyzed as an adaptive trial using both Bayesian decision-theoretic and a frequentist group-sequential approach incorporating interim analyses after every 36 patients. Endpoint was chemotherapy treatment modifications (any vs. none). Bayesian analyses were performed for different continuation thresholds and settings with and without arm dropping and both in a 'pick-the-winner' and a 'pick-all-treatments-superior-to-control' setting.

RESULTS

Treatment modifications occurred in 34% of patients in UC and OncoMove vs. 12% in OnTrack (P = 0.002). Using a Bayesian-adaptive decision-theoretic design, OnTrack was identified as most effective after 72 patients in the 'pick-the-winner' setting and after 72-180 patients in the 'pick-all-treatments-superior-to-control' setting. In a frequentist setting, the trial would have been stopped after 180 patients, and the proportion of patients with treatment modifications was significantly lower for OnTrack than UC.

CONCLUSION

A Bayesian-adaptive decision-theoretic approach substantially reduced the sample size required for this three-arm exercise trial, especially in the 'pick-the-winner' setting.

Precision Oncology-The Quest for Evidence

The molecular characterization of patient tumors provides a rational and highly promising approach for guiding oncologists in treatment decision-making. Notwithstanding, genomic medicine still remains in its infancy, with innovators and early adopters continuing to carry a significant portion of the clinical and financial risk. Numerous innovative precision oncology trials have emerged globally to address the associated need for evidence of clinical utility. These studies seek to capitalize on the power of predictive biomarkers and/or treatment decision support analytics, to expeditiously and cost-effectively demonstrate the positive impact of these technologies on drug resistance/response, patient survival, and/or quality of life. Here, we discuss the molecular foundations of these approaches and highlight the diversity of innovative trial strategies that are capitalizing on this emergent knowledge. We conclude that, as increasing volumes of clinico-molecular outcomes data become available, in future, we will begin to transition away from expert systems for treatment decision support (TDS), towards the power of AI-assisted TDS-an evolution that may truly revolutionize the nature and success of cancer patient care.

Tumor molecular profiling of NSCLC patients using next generation sequencing

Non‑small cell lung cancer (NSCLC) is the most common type of lung cancer and a tumor with a broad spectrum of targeted therapies already available or in clinical trials. Thus, molecular characterization of the tumor using next generation sequencing (NGS) technology, has become a key tool for facilitating treatment decisions and the clinical management of NSCLC patients. The performance of a custom 23 gene multiplex amplification hot spot panel, based on Ion AmpliSeq™ technology, was evaluated for the analysis of tumor DNA extracted from formalin-fixed and paraffin-embedded (FFPE) tissues. Furthermore, the Ion AmpliSeq™ RNA Fusion Lung Cancer Research Panel was used for fusion RNA transcript analysis. The mutation spectrum of the tumors was determined in a cohort of 502 patients with NSCLC using the aforementioned targeted gene panels. The panel used for tumor DNA analysis in this study exhibited high rates (100%) of sensitivity, specificity and reproducibility at a mutation allelic frequency of 3%. At least one DNA mutation was detected in 374 patients (74.5%) and an RNA fusion was identified in 16 patients, (3.2%). In total, alterations in a cancer-driver gene were identified (including point mutations, gene rearrangements and MET amplifications) in 77.6% of the tumors tested. Among the NSCLC patients, 23% presented a mutation in a gene associated with approved or emerging targeted therapy. More specifically, 13.5% (68/502) presented a mutation in a gene with approved targeted therapy (EGFR, ALK, ROS1) and 9.4% (47/502) had an alteration in a gene related to emerging targeted therapies (ERBB2, BRAF, MET and RET). Furthermore, 51.6% of the patients had a mutation in a gene that could be related to an off label therapy or indicative for access to a clinical trial. Thus, the targeted NGS panel used in this study is a reliable approach for tumor molecular profiling and can be applied in personalized treatment decision making for NSCLC patients.

Precision oncology: neither a silver bullet nor a dream

Precision oncology is not an illusion, nor is it the magic bullet that will eradicate all cancers. Precision oncology is simply another weapon in our growing armament against cancer. Rather than honing in on the failures of a relatively young field, one should advocate for integrating its successes into widespread clinical practice, especially for indications, such as: ABL, ALK, BRAF, BRCA1, BRCA2, EGFR, KIT, KRAS, PDGFRA, PDGFRB, ROS1, BCR-ABL, FLT3 and ROS1, where aberrations have been shown to alter responses to US FDA approved drugs - that is, level 1 data. Moreover, to truly assess the promise of precision oncology, we must first begin by defining our expectations for this field. Importantly, we must recognize that the conception of precision oncology arose as an antithesis of the 'one-size fits all' cancer therapeutics approach. Consequently, tools used for evaluating these conventional, large-scale trials, are not directly transferable for assessing nonconventional, smaller-scale trials needed for evaluating precision oncology. Hence, a thorough vetting of precision oncology as another tool of the trade, must first begin by reassessing our expectations for this field, as well as current clinical trial designs and end point measurements. Importantly, we must recognize that most targeted therapy approaches are in their infancy, with only monotherapy approaches being assessed and combination therapies likely being necessary to fulfill the promise of precision oncology.

Ethical considerations of neuro-oncology trial design in the era of precision medicine

The field of oncology is currently undergoing a paradigm shift. Advances in the understanding of tumor biology and in tumor sequencing technology have contributed to the shift towards precision medicine, the therapeutic framework of targeting the individual oncogenic changes each tumor harbors. The success of precision medicine therapies, such as targeted kinase inhibitors and immunotherapies, in other cancers have motivated studies in brain cancers. The high specificity and cost of these therapies also encourage a shift in clinical trial design away from randomized control trials towards smaller, more exclusive early phase clinical trials. While these new trials advance the clinical application of increasingly precise and individualized therapies, their design brings ethical challenges . We review the pertinent ethical considerations for clinical trials of precision medicine in neuro-oncology and discuss methods to protect patients in this new era of trial design.