Data Collaborations Are Driving Improvements to the Use of Real-World Evidence

The number of expedited FDA approvals for cancer drugs that are based on surrogate end points is increasing, leading many to search for new ways to uncover efficacy and safety data to justify the costs associated with these treatments. With the growth of data innovations and collaborations, the answer might be found in real-world evidence (RWE), according to presentations at the 9th Annual Research Symposium held in conjunction with the AMCP Nexus 2019 meeting,

"We can't expect this to directly replicate clinical research, but it's really important to advance this methodology," said Jeff Allen, PhD, president and chief executive officer, Friends of Cancer Research (FOCR), a group that was instrumental in the development of the breakthrough therapy designation. "If we're going to get back to the idea of how we're going to expedite the development of treatments for patients who have no other therapeutic options, but also carry out the responsibility to continue to evaluate the effect of these therapies and benefits over time, this is a way to advance that need in a different way."

Since the launch of the breakthrough designation in 2012, the FDA has approved 150 breakthrough therapies and granted more than 332 breakthrough therapy designations, according to FOCR.¹ In an analysis of the impact of the breakthrough program conducted in 2016, the designation was found to accelerate premarket drug development by 2.2 years, representing a significant trend toward rapid access to potentially lifesaving medications.²

However, this new paradigm for drug approval has generated controversy, since a majority of the regulatory decisions under the program are based on surrogate end points and not on overall survival (OS). While OS is the gold standard, there are many challenges that make it difficult to achieve, even in large, randomized phase 3 trials, Allen noted. In general, cross over between arms commonly confounds the calculation of OS, and, in some cases, the amount of time needed to show a statistical benefit is simply too long.

RWE has the potential to overcome these challenges, Allen noted, and FOCR has teamed up with several data partners, including the American Society of Clinical Oncology (ASCO)'s CancerLinQ, Cota, and Flatiron Health, to conduct a pilot study to find out exactly how it can help. The analysis has shown that patients in the real-world setting were older than their randomized controlled trial counterparts and that data collected in the pilot study could be used to compare across treatment groups for several efficacy end points.

Through the collaboration, clinical end points like progression-free survival (PFS), time to next treatment, and time to discontinuation could be tracked in a patient population that closely mirrored a true treatment setting, Allen said. Many of these end points correlated with OS; however, survival itself still remained challenging to calculate, as the date of death is not always available, he added.

"What we were able to see, at least from the initial partners who did this analysis for us, is that they were able to construct, with great detail, the different characteristics of the populations who were receiving each product," Allen said. "They were also able to show that some of these non-traditional measures, like time to treatment discontinuation, actually correlated quite well with traditional measures, like overall survival."

Advancing Cancer Surveillance

The Surveillance, Epidemiology, and End Results (SEER) program, which has been funded by the National Cancer Institute (NCI) since 1973, houses real-world data for 34% of the US population. While the registry has traditionally had several limitations, the NCI is looking to expand the reach of the database through partnerships, according to Donna Rivera, PharmD, MSc, a scientific project officer at the NCI.

Through these partnerships, the NCI is looking to add data on comorbidities, adverse drug events, treatment success, second and third courses of treatment, and health outcomes, such as survival. SEER has traditionally only collected diagnosis, staging information, tumor characterization, and initial course of treatment.

"The main goal for the enhancement is to create a system that represents population-level, real-world data to supplement clinical trials and understand the effectiveness of oncology care outside the clinical setting," said Rivera. "Clinical trials represent 3% to 5% of the adult oncology population, however, that leaves 95% of patients outside of clinical trials, for which we need quality data."

The new initiative looks to add further biomarkers, through partnerships with Genomic Health and Myriad/Invitae. Additionally, partnerships with CVS and Walgreens will provide data on outpatient prescription use for oral medications, with the ability to measure not only utilization but also adherence. The collaboration will also tap into electronic health records (EHRs) through ASCO's CancerLinQ program and will include information on radiation oncology through partnerships with Varian/Elekta.

As an example of a recent approval that could benefit from additional RWE, Rivera discussed the tumor agnostic indication for larotrectinib (Vitrakvi), which was a first for the FDA. The approval for larotrectinib, which arrived in November 2018, was based on objective response rates (ORR) across 3 early phase clinical trials that enrolled 55 adult and pediatric patients with a variety of TRK fusion-positive cancers. In the studies, the ORR was 75% across groups by independent review.³ RWE could be used to provided further insight on the treatment, she said.

"The changing landscape of cancer treatment includes rapid drug development," said Rivera. "We need to be able to understand these new medications being approved, especially within the context of genomic testing, the integration of diagnostics testing into patient treatment, and increased survivorship."

Once fully in place, the enhanced SEER database will become a tool for monitoring a variety of factors, down to the most common medications prescribed for a given type of cancer by region. Additionally, the registry will hold ample data to allow for trend analyses, monitoring of patient adherence and compliance, clinical outcomes, and differences in use across subpopulations, Rivera noted.

"Pharmacists and healthcare providers now have a resultant education requisite to learn data science awareness as an integral aspect of clinical practice in the era of big data," said Rivera. "As data shapes evidence-based care, pharmacists with clinical data science awareness can expand their role as collaborators on the interdisciplinary team. Collaboration in the data science and real-world evidence space is absolutely essential."

FDA Adds Perspective

Examples of the FDA's feelings toward RWE are starting to emerge. In April 2019, the agency approved the cyclin-dependent kinase 4/6 inhibitor palbociclib (Ibrance) in combination with endocrine therapy for the treatment of men with hormone receptor-positive, HER2-negative advanced or metastatic breast cancer, based on an assessment of additional clinical data and findings from EHRs and IQVIA.

Despite being a step in the direction of incorporating real-word evidence into a regulatory decision, Harpreet Singh, MD, from the FDA, noted that the decision was based primarily on clinical data and not on the data from the EHRs or IQVIA. She highlighted several of the drawbacks inherent in currently available RWE, including a lack of full data on clinical outcomes like PFS or OS, further emphasizing the need for new models built on collaboration.

"The rationale for why we felt we were able to extrapolate and expand the indication for palbociclib to include male patients was largely based on the registration trials, PALOMA 2 and 3. We found the real-world evidence to be supportive but not driving our decision," said Singh, team leader for Breast Cancer Drug Development at FDA. "It didn't tell us anything differently from what we would have otherwise expected. There are well-characterized safety profiles for this drug. There were no new safety signals."

While the use of RWE for the approval of palbociclib signals a step toward acceptance for the FDA, there is still a long way to go. Efforts from FOCR and the NCI are steps in the direction of adding more clinical meaning to the data, which is currently largely based on claims data.

"Because of the limitations in real-world evidence, our solution at the agency is to ask sponsors to include patients in their prospective clinical trials," said Singh. "There is potential for more use of real-world evidence. We're not stopping but in this one example there certainly were limitations."

References

1. Breakthrough Therapies. Friends of Cancer Research website. https://www.focr.org/breakthrough-therapies. Accessed October 29, 2019.
2. Shea M, Ostermann L, Hohman R, et al. Impact of breakthrough therapy designation on cancer drug development. Nat Rev Drug Discov. 2016;15(3):15.
3. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of larotrectinib in TRK fusion—positive cancers in adults and children. N Engl J Med. 2018; 378:731-739. doi: 10.1056/NEJMoa1714448.