F. Randy Vogenberg: There have been attempts to look at large claim databases (eg, WellPoint's) of patients receiving erythropoiesis-stimulating agents (ESAs). News about the cutback on ESA use was accompanied by reports about the rise in the number of transfusions. At the same time, articles published in pharmacy and hospital journals were advocating the use of ESAs to avoid transfusions. So, the message is mixed. What people are doing in an inventory environment is not the same as what they are doing in an institutional inpatient environment. And hospitals may continue to use ESAs to avoid the more costly transfusion, a purely cost consideration. Is this how you see it?
Samuel Silver: I think this may be true. But the issue of transfusion is beyond cost, especially in the outpatient setting today, where 85% of cancer patient care occurs in the community and not in academic medical centers.
The first issue is that giving transfusions at centers such as the University of Michigan is relatively simple, and our facilities are well prepared to administer them. We have a good blood bank on site and plenty of chairs for transfusions, so we can get people set up, and we have blood irradiators (see below). But in the community, physicians by and large don't give transfusions in their offices because of the administrative and procedural hassle. Rather, they send patients to their hospital outpatient infusion area, which is usually very small, where they can run out of transfusion chairs relatively rapidly, so that we now have a group of patients either requiring admission for transfusion or experiencing a significant delay waiting for an outpatient chair. And these are not patients with shortness of breath who require admission but rather it's a quality-of-life (QOL) matter, because of the requirement of an inpatient stay.
The second issue is that the typical Medicare beneficiary has many comorbid conditions (eg, heart failure) in addition to cancer. Thus, we are no longer talking only about potential infectious complications of transfusion but also about issues of fluid overload that would not occur when using an ESA.
The third concern is that many patients receiving chemotherapy, especially those with hematologic malignancies, are immunosuppressed. They require irradiated blood products to avoid transfusion-associated graft-versus-host disease. Irradiating blood is very easy; any university blood bank will have a blood irradiator. It does not require much time, but it will add some expense. Nonetheless, it is well worth it. However, if the patient is in a suburban or rural community, that patient may not have immediate access to irradiated blood and will have to wait a day or two to get it. So, all of a sudden something that would be simple for me at the University of Michigan becomes a process that takes several days for a patient that receives chemotherapy in a suburban or rural facility. These are some of the noneconomic, procedural issues that are involved in giving transfusions.
Vogenberg: The other side of this issue concerns younger populations. The 50- to 60-year-old patient may still have a commercial insurance and not Medicare. Thus, there are also direct economic costs to consider, such as absence from work, particularly when the patient may be tied up for a day or two, as you said. There are many repercussions for Medicare, which is the focus for the Centers for Medicare & Medicaid Services (CMS), and for commercial insurers and employers, who are paying the bills. This is not well understood; the tendency is to just look at the drug cost and the drug companies, and this is more emotional than rational.
Silver: True. And the next thing we know, we are talking about Medicare Part B, and we begin to reach out to Medicare Part A, when we start admitting people to the hospital. Apparently those involved in Medicare Part B decisions don't see this, but we certainly see this as citizens and on the private side. And the employers see this. So the US Food and Drug Administration (FDA) has issued new warnings, which essentially reflect the concerns about potential tumor progression, venous thromboembolism, and keeping the hemoglobin level at <12 g/dL (although this cut-off measure was dropped by the FDA from the most recent ESA labeling).
The FDA reiterated this in its most recent statement. That is where we are from the point of view of the FDA and CMS. And the Senate passed a nonbinding resolution about it, since the Senate has a sense that the National Coverage Decision (NCD) has no weight and should be changed.
Vogenberg: This NCD issue probably comes at a good time for CMS, because the current Congress is deadlocked on many issues. So nothing is going to happen. There are no repercussions against CMS's decision about ESAs, and it can come back and say it has saved a lot of money, while preventing further harm to their beneficiaries. It is really almost a neutral situation from a public policy perspective.
Silver: In the meantime, Representatives Peter Stark and Henry Waxman are saying they are protecting the patients from this "miracle grow for tumors." And because we have no quality-of-life (QOL) data, there are no improvements in QOL issues. Now what do we do? We definitely require more data.
Vogenberg: How transparent is CMS regarding its data?
Silver: CMS is not transparent at all. When you read the NCD, their background materials have been tightened up a bit. But the intellectual basis to bridge between the final coverage decision background material and the rules is missing.
Vogenberg: This is very typical of what you would see in almost any insurance company. You get caught up in this dynamic of economics and cost, and you are trying to provide just enough access so that people are happy, but at the same time less attention is paid to QOL than to the science, so it really is about economics. For CMS, it appears to be about money. They had to stay within a budget, and they knew the ESA coverage was a budget buster and a highexpenditure category, as it is for any employer. This is the perfect moment to come up with a Solomon-like decision. CMS is not required to disclose as much as the FDA. This makes it easier for the commercial carriers to follow suit.
Silver: In addition, CMS will identify 10 more questions that should be answered, as they did 1 year ago, even though many of these questions are unanswerable, because the data don't exist, and CMS knows it; a true Catch-22.
Vogenberg: This points to an interesting dichotomy between what the Agency for Healthcare Research and Quality is doing and Medicare Part D initiatives and the general Medicare initiatives about quality issues in hospitals, such as pay-for-performance. On one hand, CMS is saying it is going to pay for improved performance and outcomes in the hospitals, which, based on the current literature, suggests they would be using ESAs; these drugs lead to better results and reduce complications and readmissions, which are the criteria for quality. But on the other hand, Medicare coverage decision, which was based on an outpatient basis, prevents clinicians from using these drugs.
Silver: That brings up an interesting point. In CMS's Physicians' Quality Reporting Initiative, one of the early reporting initiatives dealt with the use of ESAs in myelodysplastic syndrome (MDS). It was necessary to make sure patients had sufficient iron stores, which was a quality point. Yet, the payment for MDS associated with ESAs was denied at that time, as we discussed in the first part of this interview (AHDB. May 2008;1:46-50).
Vogenberg: And then it could be argued that the cost of these medications is too high?
Silver: Actually, Medicare is not supposed to talk about costs or to consider it in its decisions, but of course it is the underlying context.
Vogenberg: And what is the Veterans Administration (VA) doing with the coverage?
Silver: I do not have the answer to that. The VA has an excellent electronic medical records system; however, in the past few months the VA was unable to use any of their data for Health Services Research purposes because of confidentiality and HIPAA (Health Insurance Portability and Accountability Act) issues.
Vogenberg: There must have been some kind of analyses done within the VA military review structure when all this was transpiring, because they are so cost and quality conscious. I wonder what the VA system is doing about TRICARE and the active military, which represents a large population?
Silver: This is a very important issue. The FDA has asked the American Society of Clinical Oncology (ASCO) and the American Society of Hematology (ASH) to determine what studies should be done in the future. ASH has asked the American Association of Cancer Research (AACR) to participate in this initiative. We are trying to convene a meeting with representatives from the FDA, CMS, ASH, ASCO, and AACR to determine what trials are needed and who should be running them.
Vogenberg: Would they be doing a corresponding health economics analysis with clinical studies parameters?
Silver: That would be an important thing to look at as well, but the executive committee of ASH believes that the National Institutes of Health (NIH) should not be paying for such studies to help determine the best hemoglobin range for ESA use. The Executive Committee is concerned that this type of study costs a lot of money. And diverting NIH money for this clinical question would leave less money for important basic science research.
Vogenberg: That is a good point.
Silver: It would be important to engage both Johnson & Johnson and Amgen to design an open study that would be funded by an unrestricted grant from both drug manufacturers to examine the pathophysiology of ESA-associated thromboembolism. And maybe we should involve the NIH and the National Cancer Institute or the National Institute of Diabetes and Kidney Disease to address issues surrounding endstage renal disease (ESRD). So we need to focus on 3 aspects of the drugs—tumor promotion, the biology of venous thromboembolism, and the relationship to ESRD, which can teach us a lot. And these 3 things should happen in parallel, along with the economic issues that are related to these clinical considerations.
Vogenberg: Clinically speaking, where would you want the focus of these studies to remain?
Silver: Future studies should somehow focus on QOL in a meaningful way. Combining the efforts of Johnson & Johnson and Amgen may be effective. In that way, individual marketing agendas could be put aside. For at least the moment, private payers have been using the ASH/ASCO guidelines as opposed to Medicare guidelines. This, of course, could change tomorrow. When I spoke with Dr Lee Newcomer of UnitedHealthcare a while back, he made no promises. He seemed to think it was very reasonable to use the ASH/ASCO guidelines, using an initial hemoglobin level of <10 g/dL, and then maintaining patients at hemoglobin levels between 10 g/dL and 12 g/dL. As I mentioned in the previous article in this journal (May 2008), oncologists have come to view this issue with concern; we are generally maintaining patients at or below hemoglobin levels of 12 g/dL.
Vogenberg: Are physicians still providing ESAs as freely as they had before the new Medicare coverage ruling was announced?
Silver: No. In fact, the number of those using ESAs has declined significantly. This may have been associated with the FDA issuing the black box warnings in March 2007. I think the bottom line is that most physicians want to use drugs in a reasonable way, and perhaps they overcompensate when they hear the cynical perspective about perverse physician incentives. So when the black box warning was issued, there was a great deal of self-correction.
Vogenberg: Carriers will be more willing to take the soft approach UnitedHealthcare has taken, unlike that taken by Medicare. They will seek the benefit of the reduced expenditures by informing their provider networks and offering them the opportunity to avoid potential complications and to be in compliance with the guidelines. This could be a win win situation for everybody.
Silver: Possibly. But it is a complicated issue. Patients getting chemotherapy are receiving toxic agents with therapeutic ratios that are some of the most difficult in medicine. Many of the patients, especially those in the Medicare population, have incurable cancers; they are receiving chemotherapy mainly for palliation. The goal is to give them as high a QOL as possible. If they are receiving chemotherapy and spend the last 2 months of their lives in the hospital, we are not doing the right thing.
Debilitating patients for the majority of their remaining time to shrink a tumor is not good care. A Pyrrhic victory in this scenario is no victory at all. And using a drug that potentially can cause thromboembolic events only complicates this picture.
I was called to an FDA meeting in which a patient representative was anxiously going on about tumor growth associated with ESAs that patient was receiving. This is a difficult thing to say, but having progression of an incurable cancer is not always the worst thing that can happen. Chemotherapy is rarely provided with curative intent in this population. Appropriate palliation may include chemotherapy along with adjunctive therapies, such as ESAs, that may cause a small risk of tumor growth.
Vogenberg: This is a valid point that many people forget. Patients who are seriously ill often lose sight of reality or don't want to face it.
Silver: Right. Sometimes we need to reach deeper into the bag for more information. As I have said, finding this would be tantamount to finding the Holy Grail, and unfortunately, you cannot infallibly predict outcomes.
Dr Silver is Professor of Internal Medicine, Director, Cancer Center Network, Division of Hematology/Oncology, and Assistant Dean for Research, University of Michigan Health Systems, Ann Arbor, MI.
Disclosure Statement Dr Silver is a consultant to Bear Stearns, Lehman Brothers, and the Gerson Lehrman Group, and receives grant/research support from Blue Cross Blue Shield of MI.