Medicare would cover sipuleucel-T for Food and Drug Administration–approved uses, the Centers for Medicare and Medicaid Services proposed in a draft national coverage decision released March 30.
Off-label use of the novel therapeutic prostate cancer vaccine, marketed by Dendreon under the trade name Provenge, would be left to the discretion of local Medicare claims carriers, the agency said.
The draft national coverage determination (NCD) would not change the reimbursement situation for on-label uses for Provenge. As of early January, all 15 Medicare subcontractors in the United States were covering the vaccine for its approved indications, according to Dendreon. Provenge was approved by the FDA in April 2010 for patients with asymptomatic or minimally symptomatic, metastatic, castrate-resistant (hormone-refractory) prostate cancer.
For off-label uses, the draft NCD listed a number of reasons why the agency decided against prohibiting Medicare coverage. First, because there is currently no evidence to support additional indications, the CMS said that it is "hopeful that unlabeled uses in the near future will take place only in the context of bona fide clinical studies."
However, "if this turns out to be an overly optimistic viewpoint," according to the draft, the agency may "reconsider this NCD to ensure that Medicare coverage is restricted to uses that are supported by robust evidence."
Other reasons for not barring off-label reimbursement relate to providing contractors with the flexibility to quickly adapt coverage requirements as physicians gain experience with the vaccine, according to the draft decision.
Because the "potency of the administered preparation itself is unique from patient to patient and may even differ in a given patient from administration to administration," future evidence may demonstrate "improvement in patient health outcomes as a result of incremental improvements to the current labeled process," according to the draft decision. The process of administration involves autotransfusion of the patient’s own treated white blood cells.
Should future evidence demonstrate improvements in patients who do not meet the current labeled indication, "we want our administrative contractors to have the flexibility to determine local coverage without the need to reconsider an NCD."
The CMS noted that it considered requiring that coverage for Provenge be premised on the collection of outcomes evidence (using "coverage with evidence development" provisions), but decided against that course because the data are already strong enough to support on-label uses and there is no evidence to support uses off label.
Panelists on the Medicare Evidence Development and Coverage Advisory Committee urged further data collection for Provenge during a November meeting. A number of members suggested building on the patient registry that Dendreon will establish to monitor Provenge safety as a way to collect information that can fill some of the evidence gaps.
The Provenge registry, which will include a minimum of 1,500 patients, was originally intended to track incidences of cerebrovascular adverse events linked to treatment; it is part of a REMS (Risk Evaluation and Mitigation Strategy) required by the FDA as a condition of approval.
Following the MedCAC meeting and a subsequent meeting with the CMS, Dendreon committed to collecting additional data in the registry.
A large proportion of patients in the registry are expected to be on Medicare, Dendreon said. As a result, "there will be the opportunity ... to compare outcomes across age groups, and also to compare the survival of individuals in the registry stratified by age to the outcomes of similar patients in the pivotal studies of Provenge and other publicly available databases."
The registry will support further analysis of whether black men may benefit more than other groups from Provenge, which was suggested in the pivotal clinical studies on the therapy, the company added. Some 5%-10% of treated patients are expected to be black.
"The findings will be considered both in comparison to the pooled survival estimate from the pivotal trials in this sub-group, and to the survival of subjects of other races in the registry. Each of these comparisons may lend further insight into the question of whether Provenge is substantially more beneficial in this or any other sub-group than it is in the general population."
The company will also catalog the type and timing of therapies that patients receive prior to and subsequent to Provenge.
"We will include these treatments as covariates in models of overall survival, allowing us to generate hypotheses about potentially positive or negative interactions between Provenge and such therapies," according to the company.
The CMS will accept comments on the draft guidance for 30 days and release a final national coverage decision by June 30.