Could Healthcare Reform Speed Molecular Medicine Adoption?

Jason J. Luke, MD, FACP

Editor-in-Chief of Contemporary Oncology

Instructor, Department of Medicine, Harvard Medical School Staff Physician, Medical Oncology - Melanoma, Dana-Farber Cancer Institute

Over the past decade drug development in cancer has shifted to an increasingly gene target—based approach.

The development of drugs such as erlotinib for EGFR mutant non-small cell lung cancer (NSCLC) and vemurafenib for BRAF mutant melanoma have established the power of targeting genomically selected patient populations. A product of this success, however, is that most of the common (and druggable) genomic aberrations in cancer have now been targeted. While the power of targeted approaches remains, the subsets of patients harboring new targets is shrinking. For example, major clinical responses have been observed in urothelial cancers harboring FGFR alterations, but the frequency of these genomic changes is relatively low (10%-20%), in a relatively infrequent cancer (<12,000 cases of metastatic disease per year in the United States).¹

Similarly, inhibition of RET fusions in NSCLC appears to be a promising target; however, the incidence of these genetic abnormalities is only on the order of 1% of NSCLC.²

This is even more worrisome in the context of everyday practice where genomic data are increasingly becoming available to patients and physicians. Many of these data remain both uninterpretable and irrelevant to the practicing physician. However, it has become common when reviewing results from genomic testing to find rare mutations that do seem to be actionable but for which there is no drug immediately available. The urgency of this problem has become more pressing given the development of research around “exceptional responders,” such as a complete response to the US Food and Drug Administration (FDA) approved medicine everolimus, in a patient with TSC2-mutant thyroid cancer.³

At the same time that our genomic approaches are expanding, US healthcare is changing a great deal in the context of the reforms outlined in the Affordable Care Act (ACA) of 2010. The ACA is driving providers to organize systems around Accountable Care Organizations (ACOs) and appears to be on the verge of shifting to a bundled payment strategy. Additionally, there is a push for transition of all health records to electronic health record formats, as well as for patient data sharing across platforms. These changes are difficult in many circumstances but do provide for a framework that could improve drug development and patient access to targeted therapies in the future.

As providers and health systems consolidate for cost savings and organizational purposes, incorporation of cancer genomic testing for all patients is a goal that could become a reality. With economies of scale, incorporation of molecular testing and genomics could be included in the bundled payment and the size of the growing health networks could put price pressure on the cost of genomic assays. The electronic medical records of these larger health systems may have a greater potential to centralize and protect sensitive patient genomic information. Additionally, the connection of oncology practices throughout health networks could facilitate the development of virtual tumor boards for decision making regarding molecular testing and best practices in the application of targeted therapies for cancer.

The impact of these changes on patient care could be significant. If all patients underwent molecular profiling, it would likely be much easier to identify patients with rare or unusual genomic lesions. These patients could then be directed toward molecularly directed treatment options before receiving what are often ineffective second- and third-line therapies. Such changes could also improve access to more effective medicines in remote areas or for rare genomic alterations that would otherwise go untreated.

The upcoming changes to the health system could also have a major effect on drug development for rare molecular lesions. Scenarios could be imagined in which pharmaceutical companies contract with cooperative groups or large health networks in order to find the rare patients needed to validate a new targeted therapy. In some ways there are signs that the drug development community is embracing this direction already.

The NCI MATCH trial is an example of this.⁴ The NCI MATCH Trial is a master protocol in which several small phase II trials of targeted drugs will be administered to patients of different cancer histologies who harbor specific molecular aberrations and are located at medical centers across the country. A focus of MATCH will be on rare cancers—at least 25% of patients enrolled are expected to fill this definition. Industry is also showing signs of heading in this direction with the development of the Novartis Signature program⁵ as well as a Genentech multi-arm trial evaluating HER2, BRAF, EGFR, and Hedgehog pathway activation in patients for whom there is no approved targeted therapy.⁶ If the field continues to move in this direction, a transition may take place in which phase I trials could be completed quickly for safety and pharmacokinetics and then larger, adequately powered phase II studies could be rapidly accrued to assess the efficacy of the targeted drugs in specific molecular subsets.

Many barriers to such a future of drug development do exist. There is an obvious requirement for adequate protections to be put into place around sensitive genomic patient data in large electronic health networks. A drug development paradigm such as that described above would require collaboration between industry, the National Cancer Institute, cooperative groups, and large ACOs. Perhaps most difficult, the FDA would need to facilitate a drug approval process that could be simplified so that aspects of it could be completed outside of specialized or academic centers.

It is clear that there is great promise in targeted therapeutics for cancer and that the environment for healthcare delivery is rapidly changing. While many oppose some of the changes that have been proposed, there are also potential upsides that the medical community and those working in drug development could harness. I hope any of us in the cancer community who have influence will do their best to optimize the opportunities to bring these 2 changing landscapes together.

References

1. Sequist LV, Cassier P, Varga A, et al. Phase I study of BGJ398, a selective pan-FGFR inhibitor in genetically preselected advanced solid tumors. Abstract CT326. Presented at the American Association for Cancer Research Annual Meeting 2014. Accessible at: http://bit.ly/1Dwq72l.
2. Drilon A, Wang L, Hasanovic A, et al. Response to cabozantinib in patients with RET fusion-positive lung adenocarcinomas. Cancer Discov. 2013;3:630-635.
3. Wagle N, Grabiner BC, Van Allen EM, Amin-Mansour A, Taylor- Weiner A, Rosenberg M. Response and acquired resistance to everolimus in anaplastic thyroid cancer. N Engl J Med. 2014;9;371: 1426-1433.
4. National Cancer Institute. NCI molecular analysis for therapy choice program (MATCH) & pediatric MATCH. http://1.usa. gov/1vgzxeb.
5. Novartis oncology. About signature. https://www.signaturetrial. com/en.
6. Clinicaltrials.gov. A study evaluating Herceptin/Perjeta, Tarceva, Zelbroraf, and Erivedge treatment targeted against certain mutations in cancer patients. ClinicalTrials.gov identifier: NCT02091141. http://1.usa.gov/1tDjZEj.