CMS has abandoned the idea of limiting some uses of erythropoiesis-stimulating agents to a research trial setting, despite receiving public comments that the "coverage with evidence development" approach may be acceptable in some circumstances.

The agency's July 30 final national coverage determination on ESAs for non-renal disease uses, primarily cancer-related uses, draws some clear boundaries on when Part B claims contractors can pay for ESAs.

The changes from the proposed NCD are a mixed outcome in terms of specific uses of ESAs. For example, the agency made final its policy that several uses cannot be reimbursed, because they fail to meet the statutory criteria that services to Medicare beneficiaries must be "reasonable and necessary." But the agency winnowed those uncovered uses from 13 to eight. CMS also modified or clarified some of the constraints, such as dosing limits, on uses that it believed could be covered.

CMS had asked for perspectives on whether the noncovered uses should be considered for reimbursement under a CED policy.

The agency did not address the CED issue directly in the final NCD, but did so within the context of reviewing comments on the noncovered conditions (¹ Also see "CMS May Take Studious Approach To Covering ESAs In Cancer Setting" - Pink Sheet, 21 May, 2007., p. 3).

For example, CMS said that all commenters who addressed it disagreed with the agency's proposal to list as an uncovered ESA use patients whose treatment regimens include monoclonal/polyclonal antibodies directed against the epidermal growth factor receptor. "However, many supported CED in lieu of noncoverage."

CMS responded that the agency has "strongly considered, as many commenters suggested, whether this indication would be appropriate for CED. However, CED restricts coverage to within research studies. Coverage would not be available to any patient outside a study. We have considered options that would enroll beneficiaries initially into observational studies that could be used to assist in designing the appropriate randomized trial. However the complexities of this option exceed the agency's current ability to manage those vastly different studies."

The agency's final decision was to withdraw this use from the list of noncovered conditions (see chart).

CED also was suggested in lieu of disallowing coverage of ESAs in patients with treatment regimens that include antiangiogenic drugs such as bevacizumub. CMS similarly withdrew this item from the noncovered list.

Genentech was critical of this proposed restriction, particularly the citation of one specific product - bevacizumub, that is, the firm's Avastin . The company said the proposal could restrict even proper uses of the cancer drug (² Also see "Avastin Use May Be Overly Restricted By Medicare ESA Proposal – Genentech" - Pink Sheet, 25 June, 2007., p. 27).

For those suggesting that CED be a vehicle for collecting data on anemia of cancer not related to cancer treatment, CMS responded by stating that in this case, "there is evidence of harm and thus we do not believe CED is appropriate for ESA use for this indication." That use was retained as a noncovered condition.

The NCD process began in March after FDA issued a public health advisory about labeling changes, including a new "black box" warning, concerning aggressive dosing of ESAs. The Medicare proposal was issued days after FDA's Oncologic Drugs Advisory Committee unanimously recommended that FDA conduct additional safety studies of ESAs to support use among cancer patients (³ Also see "EPO Dosing Questions Left Unanswered; More Data, More Meetings Needed" - Pink Sheet, 21 May, 2007., p. 6).

Many commenters contended CMS was adopting restrictive policies too far in advance of FDA's review.

Outside of their original indication for end-stage renal disease patients, ESAs, which include Amgen's Aranesp (darbepoetin alfa) and Epogen (epoetin alfa) and Johnson & Johnson's Procrit (epoetin alfa), are approved to prevent or reduce the need for transfusion in cancer patients receiving chemotherapy, for use in HIV patients receiving Zidovudine (azidothymidine) and to reduce the need for transfusions in patients scheduled for major surgery.

After receiving more than 2,600 comments, most of which challenged at least parts of the proposal, CMS scaled back those restrictions in a final NCD.

Among changes from the list of noncovered uses, a CMS press release specifically noted, "The final NCD no longer distinguishes between those cancers that have erythropoietin receptors and cancers without such receptors."

In addition, "CMS has made no determination regarding ESA use for myelodyplastic syndrome. MDS is a premalignant syndrome that transforms into acute myeloid leukemia in many patients."

Blood Supply Adequacy Not A Coverage Factor

CMS also altered guidelines for treatment and reimbursement in areas such as anemia secondary to myelosuppressive anticancer chemotherapy in solid tumors, multiple myeloma, lymphoma, and lymphocytic leukemia.

According to the final NCD, for patients with anemia secondary to anticancer chemotherapy, ESAs are appropriate when the hemoglobin is less than 10 g/dL (hematocrit less than 30%), raising the level from 9 g/dL in the proposed NCD.

(The proposal did have provisions for beginning treatment when levels fell below 10 g/dL/30 percent for patients with documented symptomatic ischemic disease that cannot be treated with blood transfusion.)

In making the change, CMS explains that it agrees with public comments suggesting that "a starting level of 9 g/dL has the potential to result in more hemoglobin dropping to transfusion levels and will thus modify our proposed decision."

However, CMS says many larger policy concerns raised by commenters about a potential for increased need for transfusions did not play into the final NCD.

"The concern about the adequacy of the nation's blood supply is not a relevant factor for consideration in this national coverage determination," the agency said. "Our focus is whether the use of ESA is reasonable and necessary to treat a particular illness."

CMS sets the maximum dose for the first four weeks of ESA therapy at 1,800 U/kg for epoetin and 9 mcg/kg for darbepoetin alfa. If, after the first four weeks, hemoglobin levels exceed 10 g/dL (hematocrit greater than 30 percent), ESA treatment is no longer covered.

However, treatment may resume at the same dose if hemoglobin falls below 10 g/dL and the dose may be increased by 25 percent if the patient's hemoglobin level rises by less than 1 g/dL and the hemoglobin level remains below10 g/dL.

Treatment should be discontinued if the hemoglobin increase remains less than 1 g/dL at eight weeks after the increased dose was provided.

Treatment Duration Limits Modified

CMS also altered the maximum duration of ESA treatment in the final coverage policy.

Under the proposed NCD, treatment would have been limited to 12 weeks per year.

In the final NCD, the agency says its intent was "not clearly understood." Now, patients may receive eight weeks of ESA therapy "following the final dose of myelosuppressive chemotherapy in a chemotherapy regimen. There are no restrictions on chemotherapy regimen frequency or duration in this decision." Thus, chemotherapy patients could receive multiple eight-week sessions of ESAs per year.

One condition removed from the treatment guidelines for covered uses is patients with an increase in fluid retention or weight (5 kg) after two weeks of treatment.

"We remain concerned that ESAs may precipitate edema and heart failure," the agency remarks. "However, weight changes in cancer patients may have a multitude of causes. [In] thrombotic events, it is typically not clear to practitioners that edema and heart failure would be due to ESA versus other causes. Therefore we will not continue this restriction."

Amgen highlighted both positive and negative aspects of the NCD in a July 31 statement.

"Based on Amgen's preliminary review, it appears that CMS has adopted a policy that will limit the availability of these vital medicines to Medicare beneficiaries with cancer," Amgen said. "While the decision makes several positive changes from the earlier proposed NCD, ESA treatment is not covered if the patient's hemoglobin level is greater than 10 g/dL."

The pharmaceutical manufacturer reiterates its view that the restrictions have "no scientific basis and are incompatible with good clinical practice."

Characterizing the restrictions as "unreasonable, impractical and unworkable," Amgen asserts that CMS "has undermined the ability of physicians to decide how best to administer ESA therapy to their patients through carefully defined dosing guidance articulated by the FDA."

Amgen endorsed some of the final policies, such as ESA treatment for multiple rounds of chemotherapy, dropping the bar on concomitant use with antiangiogenic drugs, and permitting ESA coverage for myelodysplastic syndromes based on decisions by local carriers.

Similarly, J&J said, "The NCD directs physicians to use Procrit in a manner that has never been studied in controlled clinical trials. We will continue our efforts with CMS to identify and minimize potential unintended clinical consequences of this decision for Medicare beneficiaries."

Financial firm Baird in a July 31 research note reiterated its cautious stance on Amgen shares despite "incrementally positive" final NCD. "While the language is more positive than that proposed last May, it is decidedly more negative than what we believe to be standard of care today, and believe private payers may yet have an open window for ESA restriction policies of their own."

Blue Shield of California recently adopted ESA coverage guidelines that J&J described as having "some of the more stringent reimbursement guidelines that CMS has suggested" in its May 14 proposed NCD (⁴ Also see "J&J Cautious On Sales Projections Pending FDA, CMS Decisions On ESAs" - Pink Sheet, 23 July, 2007., p. 3).

- Gregory Twachtman (g.twachtman@elsevier.com)

Medicare ESA Coverage Will Not Experiment With Clinical Trial Approach

More from Archive

More from Pink Sheet