Immunotherapy Needs a New Math

Howard Kaufman, MD

A group of doctors and other healthcare industry professionals have set out to develop a more efficient tool for assessing the true value of immuno-oncology (I/O) drugs. They note that these drugs often come with high prices that may distract from their advantages over other types of therapy. For example, Kroger Pharmacy is selling the checkpoint inhibitor ipilimumab (Yervoy) for $140 per mg. At the recommended dose of 3 mg/kg for melanoma patients, the total expense can be high. However, ipilimumab is one of the class of I/O drugs that have improved expectations on supportive care costs and survival benefit. The old measures of value may not apply. Therefore, how does one determine whether $140/mg is a fair price for the drug?

A Summary of Concerns Has Been Drafted

Howard Kaufman, MD, chief surgical officer and associate director for clinical science, Rutgers Cancer Institute, helped to organize a Value Summit on the issue at the November 2016 annual meeting of the Society for Immunotherapy of Cancer (SITC). Those who participated have now developed an action plan to improve the valuation of I/O drugs. Kaufman said that although currently available oncology drug valuation tools do not fully account for the unique benefits of I/O drugs, organizers hope to build on the capabilities of existing tools rather than “reinvent the wheel.”The group has already drafted a manuscript that summarizes the conclusions that were drawn at the SITC conference. This will be submitted for publication in an upcoming issue of the Journal for ImmunoTherapy of Cancer, SITC’s peer-reviewed publication. The value summit brought together not just oncologists, but also patient advocates, pharma representatives, drug makers, and payers. The group decided to open a dialogue with groups that have already developed drug valuation tools, engage the assistance of one or more health economists, and decide whether to initiate further research. In addition, Kaufman said, “We’re hoping to develop a perspective piece for The New England Journal of Medicine in terms of highlighting why we need to have a separate value platform developed for immunotherapy.”

Kaufman said talks are underway with ASCO and that an invitation for discussion will be extended to the Institute for Clinical and Economic Review (ICER). ASCO has developed a prototype valuation tool that can evaluate the relative merits of drugs that are compared in the same trial. This model takes cost into account but has limitations that affect its utility in patient-doctor discussions about affordability. ICER has developed a system for use by payers in their negotiations with drug makers over what to pay for drugs, but the system has drawn criticism because of its emphasis on meeting payers’ needs rather than those of other stakeholders. Doctors at Memorial Sloan Kettering Cancer Center have developed a rudimentary tool that purportedly shows whether oncology drugs are priced fairly. “It’s possible that we could get working with one of these organizations to push the value piece and take one of the frameworks and add in some things that are specific to immunotherapy,” Kaufman said. However, they may also decide to develop an entirely new system of valuation.

Kaufman said none of the existing tools has exactly what is needed to address the unique benefits and risks of I/O drugs in a cost context. “There are advantages and disadvantages of each,” he said. For example, the National Comprehensive Cancer Network (NCCN) provides an affordability rating on a scale of 1 to 5 with its Evidence Block tools for evaluating therapy choices, but “I don’t think they are geared to considering the cost issue on a larger level. I think it’s one thing to put the drug costs in, but that’s not necessarily the only cost associated with treatment.”

The NCCN Defends its Evidence Blocks

There is also what Kaufman described as amortized cost, which is additional expense related to follow-up care. In first-line melanoma treatment, patients on I/O drugs are getting long-term survival benefits and don’t need to continue with tyrosine kinase inhibitors or chemotherapy. “We need some economic understanding to figure out how to best incorporate this,” he said.Although the SITC group sees a need for improvement in I/O drug valuation, NCCN officials stated their Evidence Blocks embody sufficient flexibility for doctors to make adequate value determinations insofar as I/O drugs are concerned. “The Evidence Blocks are uniquely situated to consider the different profiles of these agents individually and in the context of traditional systemic therapies,” said Robert W. Carlson, MD, CEO of NCCN, and Joan McClure, MS, senior vice president of clinical information and publications for NCCN, in a statement.

Created by a panel of experts, the Evidence Blocks incorporate “the acquisition cost of the agent, required supportive therapy, the likelihood and expense of managing adverse events (AEs), and site of care, among other factors.” Carlson and McClure said the affordability metric does not specifically weigh costs versus benefits or costs versus toxicity; however, they said the Evidence Blocks have high utility for considering multiple measures of value at one time, including “efficacy, safety, data quality, data consistency, and affordability.”

The Duration of Treatment Could Be Shortened

The SITC group also aims to determine whether alternative means are possible for compensating manufacturers for their research and development costs; the current system incorporates these expenses into the drug price. There’s also a concern that the true AE profiles of these newer drugs may not be as mild as they appear, and Kaufman said his group would like to see more exploration of this issue. For example, the results of a study presented at the European Society of Medical Oncology 2016 Congress suggested that clinical trials of I/O drugs often fail to report treatment-related AEs. The researchers gave low marks to 90% of 81 separate trials of targeted and I/O drugs for the quality of reporting of the duration of AEs and recurrent or late toxicities. The trials were of therapies approved by the FDA between 2000 and 2015. “With some of these therapies, the toxicity might not be as low as we think it is,” Kaufman said. “We do have to compile longer-term follow-up data. If you think about it, some of these drugs are pretty new. When do you use PD-L1 versus PD-1? It looks like some of these agents might have long-term toxicities.” More long-term data would help to clear this up, he said.Similarly, more information is needed to determine how long patients need to stay on I/O drugs. Kaufman said oncologists are learning that with some of these drugs, the indicated duration of treatment may be much longer than necessary, meaning that additional savings could be realized by shortening the duration of treatment. In addition, combination therapies involving I/O drugs offer savings that should be incorporated into valuation models. This is because patients on combination therapy may receive smaller doses of particular drugs than those on single-drug regimens. “There are some interesting parameters that one could use as a basis to develop regimens that make more sense. The combinations may be much more effective. They may actually be cheaper, but then the toxicity question comes in,” Kaufman said.

Carlson and McClure say that a lot of these questions eventually will be answered. “Over time and with more experience with these agents, all stakeholders will gain a better understanding of the efficacy of these agents and which patients are most likely to benefit. Clinicians will also gain expertise in recognition and management of toxicities which will allow them to be addressed in a timely manner with appropriate therapies. All of this will ensure that these agents are used appropriately, safely, and efficiently,” they said in their statement.

Bossi P, Botta L, Bironzo P, et al. Systematic review of adverse events reporting in clinical trials leading to approval of targeted therapy and immunotherapy. Ann Oncol. 2016;27(suppl-6):320P. https://doi.org/10.1093/annonc/mdw366.05.