Friday, October 20, 2006

Guidelines in Whose Interest? - Amgen, Epoetin, Medicare, and Chronic Kidney Disease

A number of articles that have come out over the last two months raise questions about the increasingly aggressive use of epoetin for patients with anemia on dialysis for chronic renal failure. Patients with chronic kidney disease often have severe anemia. Epoetin can treat such anemia. Epoetin is manufactured under the trade name Epogen by Amgen Inc. Epoetin is usually dosed to achieve a given target level of hematocrit (Hct) or hemoglobin (Hb). The question is how much epoetin to use to reach what level of Hct or Hb.

The most recent update of the Cochrane review of epoetin in this setting concluded that more aggressive treatment attempting to achieve a nearly normal hemoglobin was not clearly supported by existing evidence from randomized controlled trials, "There was no significant difference in the risk of death for low (<>133 g/L). Lower targets were significantly associated with an increased risk for seizures but a reduced risk of hypertension." (See Strippoli GFM, Navaneethan SD, Craig JC. Haemoglobin and haematocrit targets for the anaemia of chronic kidney disease. Cochrane Database of Systematic Reviews 2006, Issue 4. Art. No.: CD003967.)

An article in Health Affairs (Cotter D, Thamaer M, Narasimhan K et al. Translating epoetin research into practice: the role of government and the use of scientific evidence. Health Affairs 2006; 25: 1249-1259 ) noted that despite this lack of evidence for more aggressive use of epoetin, Medicare, which pays for most epoetin use for patients on chronic dialysis in the US, has progressively agreed to pay for higher and higher hemoglobin targets. This policy is not based on evidence from randomized controlled trials, as noted above, nor on the US Food and Drug Administration (FDA) suggested targets for hemoglobin.

So, "spending for epoetin therapy is now the single largest Medicare drug expenditure ($1.75 billion in 2005) and is the second-largest source of dialysis facility income (approximately 22 percent)."

Cotter et al alleged that Medicare based its decisions on published practice guidelines, in particular, the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines, " in numerous policy instructions, the CMS has cited KDOQI’s target hematocrit recommendations as the basis, in part, for its epoetin Medicare policy." But,
Amgen sponsored the Kidney Disease Outcomes Quality Initiative (KDOQI) of the National Kidney Foundation (NKF), an assessment process that purported to be an evidence-based approach to evaluating the medical literature. However, a rigorous review of the evidence used to support KDOQI’s 2000 'Guideline 4: Target Hemoglobin/Hematocrit for Epoetin' recommendations yielded major deficiencies. Notably, the priority scores (that is, the weights) for cited studies were never published, and the evaluation techniques did not elicit 'linkages in the causal pathway between intervention and outcome.' Specifically, KDOQI 2000 used a few randomized clinical trials (RCTs), expert opinion, and numerous industry-sponsored observational studies based on Medicare claims and other administrative data that suggest, for any given snapshot of the treated population, that patients who achieved higher hematocrits tended to have better clinical outcomes. Thus, both researchers and guideline developers have interpreted the evidence of an association between treating anemia with higher epoetin doses and improved clinical outcomes as an indicator of causality.

Last month, Christopher Rowland writing in the Boston Globe suggested that there is additional evidence against the use of higher hemoglobin targets. He reported that a trial,

called CHOIR, was intended to document the benefits of larger Epogen doses. Instead, safety reviewers halted testing because participants were dying at an unexpectedly high rate.

Patients given the highest doses of Epogen suffered 16 more deaths -- 72 out of about 700 patients -- than those treated within FDA guidelines, who suffered 56 deaths. The results have not been published in an academic journal.

The CHOIR study focused on Procrit, a drug virtually identical to Epogen, and involved 1,432 subjects. It was sponsored by Procrit's manufacturer, Ortho Biotech Products LP , a unit of Johnson & Johnson . The intent was to determine whether patients had lower death rates and fewer cardiovascular ailments if their red blood cell counts were elevated to 13.5 grams per deciliter.
The CHOIR trial was not the first to show higher doses of Epogen carried risks. Fatal heart attacks, strokes, and other problems forced clinical investigators to suspend at least seven other trials using Epogen or similar drugs to treat kidney and cancer patients.
He also noted that "Epogen was the primary reason Amgen became a global powerhouse. In 2005, it earned $3.7 billion on $12.4 billion in revenue...."

Finally, a recent Forbes article critical of Amgen also reported the company's influence on the development of the guidelines which seem to have been Medicare's primary rationale for paying for aggressive use of epoetin,
Dr. Barry Straube, Medicare's chief medical officer, defends the Epogen guidelines.

But Straube admits that the Medicare decision on epoetin dosing is based on clinical habits and industry-promoted guidelines, not the results of randomized clinical controlled trials.

Some of the guidelines that Medicare relies on were set by a committee of the National Kidney Foundation, whose work was principally funded by Amgen. 'Amgen essentially purchased the pseudoscience that went into raising these hematocrit guidelines in the medical literature,' bristles Merrill Goozner, a director at the Center for Science in the Public Interest.
Thus this appears to be a second example this week of how guidelines developed with financial support of pharmaceutical or biotechnology companies may favor aggressive use of those companies' products, even in the absence of strong evidence for such use. (See recent post here.) At best, this can lead to excess costs and treatment that does patients little good, and risks harming them.

This is a reminder that just because a set of recommendations is called a guideline, or is written under the auspices of some august academic or scientific organization does not guarantee that its recommendations are based on the best evidence, or are not influenced by vested interests.

And yet guidelines remain a buzz word for those who say they want to control health care costs, and improve quality and access. And giving guidelines teeth, otherwise known as "pay for performance," (P4P) is now the most fashionable solution to the cost, quality, and access problems. (See post here.)

We need to remain very skeptical about how guidelines are developed, and whose vested interests they may serve, and very careful about giving guidelines teeth through P4P.


CL Psych said...

I'm not sure what Medicare shells out for newer antipsychotic medications (Zyprexa, Risperdal, Seroquel, etc.), but it must be several billion annually. This is analogous to the post here in that there is little evidence supporting their use over other less expensive meds. Granted, the old meds have their own problems as well, but at a price of 1-5% of the new meds, shouldn't we reconsider using the ridiculously more pricey medications? The taxpayer is getting soaked via Medicare's use of meds that are not backed by evidence.

Clinical Psychology & Psychiatry

Anonymous said...

What I find interesting is the reference to Forbes's. In following hcrenewal I have been struck by the number of references to business journals. I find it troubling that the WSJ, Boston Globe, New York Times, and the popular press, are asking the questions, and researching the answers, to issues that should be first discussed in the medical community,

Steve Lucas

Anonymous said...

Excellent post. We need guidelines on what constitutes a guideline, how systematic literature reviews are written, and what a consensus statement represents. I have seen all of these terms abused for their implications of fair use of evidence.

Then we need to make sure that such documents are assembled by experts in assessment of evidence who are agnostic about the particular subject area, and, of course, as free as possible of any financial interests.

Luigi said...

In the case of chronic kidney disease, there is published evidence which demonstrates that both agents are safe and effective. In hemodialysis patients, a starting dose conversion factor of 200 units epoetin to 1 mg darbepoetin. This conversion is based on the mass of the peptide portion of each molecule. In contrast, randomized clinical trial data, demonstrate the ratio necessary to produce equivalent effects ranges up to 280:1. Further studies submitted to CMS have demonstrated a dose-conversion ratio of up to 330:1. Clinicians are encouraged to consult package labeling for a useful dosing conversion table.

Anonymous said...

I am a hemodialysis patient. when my hematocrit gets up to 39 the doctors slash my epogen in accordance with medicare guidelines. Then for a few months I feel lousy and drained as my hct goes down. What is better? keeping someone alive for a few more years, or giving them the ability to enjoy the life for a few more months?

Anonymous said...

I am a healthcare worker, and I can empathize with hemodialysis patients. My brother was 45 years old and took hemodialysis for 2 years. Although I understand your statement regarding quality of life, I will also not administer a medication to a patient whose Hgb or Hct levels exceed the recommended FDA guidelines for administrations. This is called off-label use, and in a court of law, I could possibly held liable for causing harm to a patient for administering off-label use medicaiton. I would not intentionally cause a sudden cardiac event by administering a medication that carries an FDA black-box warning to a patient with a high Hgb or Hct level.