Orlistat reduced progression to type 2 diabetes over four years in the XENDOS trial

The four-year XENDOS trial found that people with obesity who took orlistat three times a day along with diet counseling lost more weight and were less likely to progress to type 2 diabetes than people who received counseling alone ^[2].

XENDOS enrolled 3,305 people aged 30 to 60 with a body mass index of at least 30 ^[2]. Their average age was 43, average weight 242 pounds and average BMI 37 ^[2]. Everyone was prescribed a diet with an 800-calorie daily deficit ^[2]. Of them, 1,640 took orlistat 120 mg three times a day and 1,637 took a placebo ^[2]. A two-hour oral glucose tolerance test was performed every six months ^[2]. At the start, average fasting blood sugar was 83 mg/dl and 21 percent of participants had impaired glucose tolerance ^[2]. People with diabetes, cardiovascular disease or gastrointestinal disease were excluded ^[2].

Over four years, 6.2 percent of people taking orlistat developed type 2 diabetes, compared with 9.0 percent of those on placebo, a hazard ratio of 0.63 (95 percent confidence interval 0.46 to 0.86, p=0.0032) ^[2]. Average weight loss at four years was 12.8 pounds with orlistat and 6.6 pounds with placebo ^[2]. LDL cholesterol fell 12.8 percent from baseline in the orlistat group and 5.1 percent in the placebo group ^[2]. Dropout rates were 48 percent with orlistat and 66 percent with placebo ^[2].

The diabetes difference showed up in one subgroup. Among participants who already had impaired glucose tolerance, orlistat reduced progression; among those with normal blood sugar, the number of new diabetes cases was too low to detect a significant difference ^[2]. Weight loss was similar in both subgroups ^[2].

Gastrointestinal side effects were common and were the main tolerability issue. During the first year, 91 percent of orlistat users reported GI side effects versus 65 percent of placebo users; by year four the rates were 36 percent and 23 percent ^[2]. In the first year, specific rates with orlistat 120 mg included oily rectal spotting (26.6 percent), flatus with discharge (23.9 percent), fecal urgency (22.1 percent), oily evacuation (20 percent) and fatty or oily stool (17.1 percent) ^[2]. These effects tend to fade quickly, with half lasting less than a week and most no more than four, and high-fat meals can make them worse ^[2]. Levels of vitamins A, D, E and K fell more with orlistat than with placebo, but the average level of each stayed within the normal reference range throughout the study ^[2].

Orlistat works differently from GLP-1 medications: it is a lipase inhibitor that blocks absorption of as much as 30 percent of dietary fat at the 120 mg dose ^[2]. It is sold as Xenical, a prescription 120 mg capsule, and as Alli, an over-the-counter 60 mg capsule ^[2]. FDA-approved labeling covers weight loss or weight maintenance in adults and adolescents 12 and older with a BMI of 30 or higher, or 27 or higher with another risk factor such as high blood pressure, diabetes or abnormal cholesterol ^[2].

Why it matters for patients

XENDOS measured a disease outcome over four years, a diagnosis of type 2 diabetes, rather than weight alone ^[2]. That is the question many people weighing a weight-loss drug actually care about: whether losing weight changes their risk of a condition, not just the number on a scale.

The results are narrower than the headline percentage suggests. The average weight difference between the two groups was about six pounds at four years, and the reduction in new diabetes cases was detectable only in people who began the trial with impaired glucose tolerance ^[2]. In participants whose blood sugar was normal at the start, the trial could not show a difference ^[2].

Tolerability is the other practical point. Nearly half of the people assigned to orlistat dropped out over four years, and 66 percent of those on placebo did ^[2], so the published rates describe those who stayed. GI effects were far more common in year one than in year four ^[2].

The trial also does not answer how orlistat compares with newer options. Whether semaglutide, tirzepatide or orforglipron produce similar four-year reductions in new diabetes diagnoses has not been established.