Behind Look Ahead
The National Institutes of Health (NIH) has announced that the Look AHEAD trial has been stopped in its 11th year, two years short of completion. The extensive trial, involving over 5,000 patients at 16 centers, was intended to find out if there was increased mortality from intentional weight loss and to see if intentional weight loss among obese patients with type 2 diabetes would result in fewer cardiovascular (CV) events. At the end of the trial, there was no difference between the study group, which received intensive behavioral counseling and the control group which received standard diabetes education and occasional support group meetings. The NIH press release indicates that both arms had lower CV rates that reported for patients with diabetes in previous studies. NIHNEWS: Weight Loss Not Reduce CV events in Type 2 diabetes
While this is news is something of a shock, many folks saw it coming. Two years into the trial, which began in 2001, the monitoring board noticed that the event rate in the control arm was much lower than expected. The expected CVD event rate in the control arm was 3.125% per year; in fact it was 0.7%. A committee was formed and made changes to the original study protocol designed to capture more events. These changes went into effect in 2008. There appeared to be three reasons for the lower event rate. First, while cardiovascular disease (CVD) is still the major cause of death in the United States, mortality has gone down, resulting from better control of dyslipidemia and high blood pressure and improved care of chronic and acute coronary syndromes. (See NCHS Data Brief, NCHS DataBrief: Prevalence of Uncontrolled Risk Factors for CVD) Second, study participants who choose to involve themselves in a long clinical trial may well be healthier than a community sample and more motivated to follow the diet and exercise and participation requirements. Finally, the Look AHEAD trial employed the Graded Exercise Test which excluded participants most likely to develop CVD. Because of the low event rate, an additional primary endpoint was added (hospitalized angina) and the trial was extended for 2 years. (See PubMed: Brancati_Midcourse Correction to clinical trial whe the event rate is underestimated: the Look Ahead Study) Readers may recall that the SCOUT trial of sibutramine also had to revise its protocol midway through the study for the same reason, resulting in a population which was older and sicker than typical clinical population. In both cases, revising the protocol did not favor the intervention.
The stopping of Look AHEAD raises a host of questions. Was the study protocol correct? Did it end up studying healthy obese diabetics? Do long-term studies produce more noise than insight? Are we really studying the aging process when we cannot control for changes in health status, drug utilization (including drugs which can increase weight) and changes in energy intake, fitness levels, etc.? What is the picture for sub-groups, such as the 60-74 age group which had good weight loss in the DPP and 4 year results of Look AHEAD? Were there specific improvements, such as reductions in medications usage, fewer hospitalizations or shorter length of stays, improvements in quality of life? Did the presence of any the alleles associated with success in bariatric surgery affect outcomes? PubMed: High allelic burden of four obesity SNPs associated with poorer wt loss. Should future efforts be devoted to cases where the disease process is already well-established or where high-risk populations can be identified and appropriate interventions evaluated? In future trials, should comorbid management be left to the local standards of care or defined in the study protocol?
Looking Ahead of Look Ahead
Whither behavioral lifestyle interventions?
The lifestyle interventions in the DPP and Look AHEAD were regarded as the ‘gold standard.’ They involved recruiting and training health professionals who provided not only the intervention but provided a supportive environment and a community spirit. Extensive communication with the patient was maintained. PUBMED: Look AHEAD: Description of the Lifestyle Intervention. Look AHEAD participants even received an honoraria of $100 at each annual visit to improve adherence. (FDA EMDAC Hearing, March 28, 2012, Dr. Rena Wing, transcript, p. 169).
Recently the CDC and the NIH were looking at ways to take the DPP/Look AHEAD model to a more replicable model. The CDC’s National DPP program awarded $6.75 million in grants to develop lifestyle interventions program among people at high risk. One involves using the YMCA to provide lifestyle counseling. http://www.ymca.net/diabetes-prevention/ Questions will certainly be asked if highly trained professionals with incentives for participants did not produce better results will a down-scale program do better?
Whither diabetes prevention?
Look AHEAD was designed following the Diabetes Prevention Program (DPP). The DPP established that both lifestyle changes and metformin could reduce the incidence of type 2 diabetes, through weight loss, although lifestyle was superior to metformin alone. Look AHEAD was taking this important finding one step further asking whether weight loss among type 2 diabetics would reduce the incidence of cardiovascular events.
Even though the DPP has been promoted as a model for preventing the development of type 2 diabetes through weight loss, there were problems.
Dr. William Knowler of the National Institute on Diabetes, Digestive, and Kidney Disease (NIDDK) told the FDA Advisory Committee earlier this year,that, after three years of the DPP,
“the rates (of development of type 2 diabetes) have tended to flatten out and become parallel among all three groups. The rate of new development of diabetes has actually slowed down in the placebo and metformin groups, compared to what it was in the first three years. And the lifestyle group has flattened out a little bit at the end, but the difference that was attained has been largely maintained over time. Notice, though, that over 10 years, although there still are remarkable treatment effects, if you look at things in an absolute sense, we can’t say that we still know how to prevent diabetes because, still, close to half of the people who enrolled in the trial developed diabetes over a 10-year period. But at least it’s been substantially delayed in those who have had the interventions.” ( FDA Endocrinologic and Metabolic Drug Advisory Committee Hearing on assessing cardiovascular safety of obesity drugs, March 28, 2012, Transcript, p 131-2).
(Figure: Diabetes Prevention Program Outcomes Study, Lancet (2009) 374; 1677-1686)
Is ‘delaying’ diabetes onset as powerful as ‘preventing’ diabetes from occurring in the competitive race for health care dollars and public attention?
Furthermore, a study earlier this year indicate poor outcomes in drug treatment of adolescents with type 2 diabetes with barely half showing glycemic control with metformin. PubMed: Clinical Trial to Maintain Glycemic Control in Youth with Type 2 Diabetes
Will the Look AHEAD experience affect FDA approval of drugs and devices to treat obesity?
The FDA has viewed obesity as a cosmetic issue and only recently acknowledged it as a disease, worthy of attention as other cardiovascular risk factors. They (meaning the FDA Endocrinologic and Metabolic Drug Advisory Committee and FDA staff) have started, just barely, to view obesity as a cardiovascular disease risk factor, like hypertension. They have also opined, from time to time, that if folks only ate less and exercised more, they would not need drugs. So how does this decision play into these views? On one hand, they may be convinced that obesity is not so easy to treat as they thought by diet and exercise. On the other hand, they may think that there is less need for anti-obesity medications because other treatments, e.g. statins, lipid-lowering drugs, anti-hypertensives, are doing their job in reducing CV risk factors. So, this view may raise the bar for approval of new anti-obesity medications. On the other (the third?) hand, we may need a re-definition of obesity which tones down its “diabetes-metabolic syndrome-mortality” axis and raises its “disability-mobility-quality of life” axis. (Running out of hands here, I would not underestimate the potential for greater evidence of obesity’s role in the development of various cancers).
The recent trend in thinking at the FDA Endocrinologic and Metabolic Drug Advisory Committee (EMDAC) has been to view anti-obesity medications narrowly as cardiovascular disease treatments. The EMDAC met on March 28th and 29th,2012 and discussed how to assess the cardiovascular benefits and safety of anti-obesity medications. At the end of March 28, Dr. Rasmussen, who is the Industry Representative on the committee had the following exchange with Dr. Eric Coleman of the FDA.
Dr. Rasmussen: In your (Dr. Colman’s) presentation, you showed that there are different populations pre-approval and in post-approval studies. ..Are we compromising the risk- benefit evaluation if we impose more risk-based patients pre-approval?
Dr. Colman: I’m not sure I understood your question. Could you rephrase it?
Dr. Rasmussen: Maybe I’m preempting some of the discussion that we’ll be having tomorrow, whether we should require more high-risk CV patients pre-approval to rule out a upper bound of the 95 percent confidence interval. But by doing so, we will likely be including older patients with established cardiovascular risk disease. And I’m wondering whether including more of those types of patients will compromise the benefits side of doing the benefit-risk evaluation.
Dr. Coleman: Yes. And it might be that if the program had the resources to do this, that that would just be one component of the program, and that there would be other be other, smaller, shorter-term studies where they could study lower-risk individuals, younger individuals for shorter periods of time.
Dr. Rasmussen: But my concern was based on the fact that the SCOUT study didn’t really – I mean, it looks like it wouldn’t be actually be able to be approved if it was submitted pre- approval. ..(FDA EMDAC, March 28, 2012, transcript at p. 334-5)
On the second day of the hearing, Dr. Rasmussen returned to the topic.
Dr. Rasmussen: So I would just like to add a little bit of perspective on what “enrichment” (Editor: “enrichment” is the term used here by the FDA referring to adding persons at high risk of CVD to the pool of subjects in obesity drug trials) in this context will mean. I mean, I did a little bit of “back-of-the-envelope” calculation, and maybe we’ll have that confirmed after lunch. But, I mean, current programs, approximately 3,000 patient-years of exposure generate 15 MACE events or so. Even if we were to double that patient-year exposure with a population of a 3-percent annual event rate coming to additional 60 events, we would still only be able to exclude a doubling of the hazard ratio. So, I mean, what we’re talking about here is actually completely shifting the population that we’re going to study in obesity programs to establish cardiovascular disease and not necessarily the population that we know actually seek treatment in the real world. So, I think that’s worth keeping in mind, that enrichment may sound appealing because it sounds like we will add a fraction of sick patients, but in reality, this will be a complete shift of the population. (FDA, EMDAC, March 29, 2012, at p. 169)
(Dr. Rasmussen’s calculations appear correction. The cardiovascular safety trial the FDA asked Orexigen Therapeutics to undertake surpassed its original goal of 7,000 patients in process of enrolling 9,000 patients to find 87 major CV events earlier than expected. Orexigen: Press Release Contrave CV study.)
A bit later, Dr. Rena Wing was asked about the influence of statins on the Look AHEAD trial. She responded:
Number one, that more and more people are being treated with statins. There’s better blood sugar control. There’s better hypertension control. So you’re going to have to look at what’s going to happen to the event rates in these studies. I was very surprised that your event rates that you’re showing me in many of these trials looked so high compared to the event rates we’re seeing in Look AHEAD. Now, some of that is because we did do GXTs. (Editor: Graded Exercise Tests.) We did select healthier patients. But I also think that if you are doing trials, in the United States especially, and with diabetics where there’s more and more emphasis on increasing the use of lipids, increasing their blood pressure control, that you’re going to be driving down your risk factors, and you’re going to have more and more confounds with medication. (FDA, EMDAC, March 29, 2012 transcript, at p. 346)
At the Cleveland Clinic’s Obesity Summit earlier this month, I asked cardiologist Steve Nissen about the FDA’s pushing companies to undertake clinical trials to rule out a CVD risk. He responded that one of the challenges of cardiovascular outcome studies of obesity drugs is that in order to get enough cv events you have to study patients with existing heart disease or at very high risk of a cv event . This pushes the trial into populations which are considerably sicker than the population likely to take the obesity drug. He suggested that FDA should look at absolute risk rather than the relative risk of the drug. If one looks at the absolute risk, you can study any reasonable population likely to take the drug. This change in the statistical approach allows one to study more typical populations.
In any event, it will be sometime before we know how the newer anti-obesity medications, like Contrave if approved), Belviq™ and Qysemia™ will impact cardiovascular disease risk factors.
Bariatric Surgery: Last Man Standing?
A study out of the Cleveland Clinic published in the New England Journal of Medicine in April, 2012 followed over 90% of 150 patients for 12 months. The study, a face to face comparison of medical therapy versus surgery in patients with uncontrolled type 2 diabetes, showed a clear superiority for bariatric surgery. The proportion of patients achieving a hemoglobin A1c level of 6% after 12 months by medical therapy alone was 12%; for those in the medical plus gastric bypass surgical group it was 42% and for the medical plus sleeve-gastrectomy group it was 37%. Weight loss was greater in the gastric bypass group (-24kg) and sleeve gastretcomy group (-25.1kg) than in the medical therapy group (-5.4kg). Use of drugs for glucose control, lipids and blood pressure control decreased in the surgical group but increased in the medical group. PubMed: Bariatric surgery versus intensive medical therapy in obese patients with diabetes
In regard to cardiovascular risk factors, a systematic review of the literature on bariatric surgery analyzed over 60 studies involving 19,543 patients. At baseline, the mean patient was 41.7 years old, female and had a BMI of 47.1. Baseline prevalence of comorbid conditions which increase risk of CVD was hypertension (44.4%), diabetes (24%) and hyperlipidemia (43.6%). After correcting for publication bias, 36% of subjects had improvements in hypertension, 26% for diabetes and 34% for hyperlipidemia. Calculating the changes for mean participants, the authors found that a woman, without baseline CVD, diabetes or smoking, who is taking anti-hypertensive drugs, will move from an 8.6% 10 year global risk for CVD to a 3.9% risk. A man, with no CVD or smoking but whose diabetes and need for anti-hypertensive drugs resolves after surgery, will move from a 10 year global risk of 18.4% to 4.7%. PubMed: Bariatric Surgery and Cardiovascular outcomes: a systematic review
So, where are we? The gold standard of lifestyle change is tarnished. The drug story is muddy at best. Bariatric surgery is clearly producing the superior results. However, access to surgery is, and will remain, a problem. The challenge for the leaders in the field is to find ways to have surgery reach more people and not be a procedure for the 1 percent. Even with greater access to surgery, the obesity-diabetes epidemic will continue to be a major health crisis. It’s time to be humble in the face of this disease and realize a lot more research is going to be needed…and soon.