Cleveland Clinic Study Finds Barriers to Precision Oncology in a Real-World Test

Davendra P. S. Sohal, MD, MPH

Assistant Professor of Medicine, Lerner College of Medicine

Director, Clinical Genomics Program

Cleveland Clinic Cancer Center

Precision oncology, the use of somatic genomic profiling to select therapies targeting molecular alterations specific to a person’s tumor, holds the promise of improving outcomes in cancer. It is a relatively new approach, made possible thanks to twin, largely synchronous advances in oncology: the widespread availability of rapid and reliable tumor sequencing, and the approval of multiple targeted therapies for clinical use.

Impact on Outcomes Established

Yet although precision oncology has promise, as proven by the study described in this article, its success hinges on the availability of clinical trials of targeted therapies.We are just now beginning to see good, prospective real-world precision oncology data to guide clinical decisions. The first major report was from a prospective clinical study on lung cancer, where tumors were sequenced for 10 common alterations, with targeted therapies available for off-label or on-trial use for most alterations.¹ The study enrolled more than 1500 patients; full genotyping was feasible in about half (n = 733, 48%). Among all enrolled patients, an actionable target was found in 466 (30%) cases, and 275 (18%) received a targeted therapy.

The key finding of this study was that when an actionable alteration was not detected, or a drug targeting a detected actionable alteration was not given (irrespective of reason), the median overall survival (OS) was 2.1 to 2.4 years. However, when a drug targeting a detected actionable alteration was given, the median OS was 3.5 years (P <.001).

Cleveland Clinic Study

This study highlighted that precision oncology can indeed improve clinical outcomes. Nonetheless, the proportion of patients benefiting was not high, despite the use of a small panel of highly actionable alterations, with off-label drugs made available to most patients with such alterations in their tumors. A similar report in breast cancer enrolled 423 patients but required fresh biopsies.² Sequencing was feasible in about 70% of cases, and showed that 195 (46%) specimens had actionable alterations. Only 43 (~10%) patients received a therapy, however.To evaluate the role of precision oncology across multiple solid tumors using a large next-generation sequencing panel, and with no up-front provision of targeted therapies—the exact situation that most oncologists find themselves in in day-to-day practice&mdash;we conducted a prospective study at the Cleveland Clinic.³

Eligibility criteria were simple: adult patients with advanced solid tumors without a known curative therapy, and an ECOG performance score of 0-2. After written informed consent, archival FFPE samples were sequenced using the FoundationOne assay. Results were reviewed at a weekly genomics tumor board including clinical and translational oncologists from various disease specialties.

Recommendations for the use of therapies targeting potentially actionable alterations, including on-label, off-label, or on-study use, were made. The treating oncologists were notified of the recommendations and followed patients for clinical decisions and outcomes.

In 2013 and 2014, we enrolled 250 patients. There were 15 tumor types, with the most common diagnoses being colorectal (25%), breast (18%), and lung (13%) cancers. Samples were shipped within a median of 7 days from informed consent, and results were available within a median of 19 days—a total turnaround time of 26 days from informed consent.

Tumor sequencing was feasible in 223 (89%) cases. The genomics tumor board review identified a potentially actionable alteration in 141 (63%) cases, and a specific therapeutic recommendation was made in 109 (49%) cases.

Findings Illustrate Hurdles

However, only 24 (11%) patients received a targeted therapy: 12 in clinical trials, 9 off-label, and 3 on-label. The most common reason for not receiving a targeted therapy was the unavailability of open clinical trials identified through public databases. Reimbursement is another hurdle as many payers do not cover next-generation sequencing panels, and off-label use of targeted therapies based on thin evidence runs counter to value-based care principles.In conclusion, we replicated the real-world setting in this prospective clinical study, and established feasibility of the process.

We learned, and continue to learn, the critical importance of cross-specialty discussions about potentially actionable alterations and best therapeutic recommendations via a dedicated histology- agnostic genomics tumor board.

This study’s findings—that tumor sequencing is feasible in many cases and that a substantial number of specimens will harbor potentially actionable alterations&mdash;are similar to those in other large reports.⁴ Ultimately, however, only a small proportion of patients actually receive therapies targeting such alterations, mainly due to lack of clinical trials, making it clear that the success of precision oncology hinges on the availability of clinical trials of targeted therapies. “Basket” studies such as NCI MATCH, Novartis Signature, and Genentech MyPathway are providing such options now.

Precision oncology continues to hold immense promise. The realization of this promise requires a concerted effort by all stakeholders in the oncology community to drive faster, reliable, reimbursable tumor sequencing; wider availability of targetedtherapy studies; continued systematic evaluation of precision oncology through histology-agnostic trials; and drug approvals and coverage based on histology-agnostic trials.

References

1. Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311(19):1998-2006.
2. Andre F, Bachelot T, Commo F, et al. Comparative genomic hybridisation array and DNA sequencing to direct treatment of metastatic breast cancer: a multicentre, prospective trial (SAFIR01/UNICANCER). Lancet Oncol. 2014;15(3):267-274.
3. Sohal DP, Rini BI, Khorana AA, et al. Prospective clinical study of precision oncology in solid tumors. J Natl Cancer Inst. 2015;108(3). pii:djv332. doi:10.1093/jnci/djv332.
4. Meric-Bernstam F, Brusco L, Shaw K, et al. Feasibility of large-scale genomic testing to facilitate enrollment onto genomically matched clinical trials. J Clin Oncol. 2015;33(25):2753-2762.