A randomised control trial to assess the clinical efficacy and metabolic effect of metformin vs pioglitazone in patients with acanthosis nigricans

Dear Editor,

Insulin resistance and hyperinsulinemia can cause hyperpigmented, thickened, velvety skin in flexural areas, known as acanthosis nigricans (AN); thus, treating these metabolic abnormalities can improve AN lesions.¹ Metformin, a biguanide, reduces hepatic glucose production and enhances insulin sensitivity, while pioglitazone, a thiazolidinedione, improves insulin sensitivity by acting on peripheral tissues.^2,3 While both metformin and pioglitazone reduce insulin resistance, their comparative efficacy in treating AN remains to be elucidated.

After approval from the institutional ethics committee, we conducted an open-label randomised controlled trial to compare the clinical and biochemical outcomes of metformin versus pioglitazone in consenting adult patients having AN aged more than 18 years. Patients on other oral hypoglycaemics with drug-induced or malignancy-associated AN, pregnant or lactating women, or those with contraindications to either drug were excluded.

Fifty patients were randomised to receive either metformin (500 mg twice daily) or pioglitazone (7.5 mg twice daily) for 12 weeks. Baseline and follow-up assessments included anthropometry, biochemical parameters (fasting insulin, glucose, triglycerides, Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), and clinical grading of AN severity and texture using Burke’s scale.⁴

Table 1 compares Burke’s grading, anthropometric, and metabolic parameters before and after metformin treatment. Table 2 presents the same comparison for pioglitazone. Figure 1 shows left axilla images taken before (1a) and after (1b) pioglitazone treatment. Table 3 compares changes between the metformin and pioglitazone groups.

Close

Figure 1:

Showing a) before and b) after treatment with pioglitazone.

Export to PPT

In terms of anthropometric outcomes, the metformin group demonstrated significant reductions in weight (2.5%) and Body mass index (2.1%). Conversely, 7 patients in the pioglitazone group experienced weight gain (mean: 1.6 kg), which has been attributed to mechanisms such as fluid retention, increased appetite, and increased subcutaneous fat deposition.⁵ Waist circumference decreased modestly in both groups (1.6% in metformin vs. 1.2% in pioglitazone) with no statistically significant inter-group difference. Sett et al. noted a significant decrease in waist circumference in both treatment arms – the metformin group and the α-lipoic acid group.⁶ Bellet et al. also noted a significant difference in waist circumference in the metformin group and the rosiglitazone group.⁷

Both medications were associated with significant clinical improvement in AN severity and texture, although no significant difference was found between the groups. As all participants had evidence of insulin resistance, this may have contributed to their responsiveness to therapy.

Biochemically, fasting insulin levels, glucose, triglycerides, and HOMA-IR showed significant improvement in both groups. Metformin resulted in a 19.9% reduction in fasting insulin, 14.3% in glucose, 9% in triglycerides, and 28.56% in HOMA-IR. Pioglitazone yielded reductions of 20.86%, 11%, 10%, and 35.8%, respectively. A Mann–Whitney U test revealed that metformin led to significantly greater reductions in weight and BMI (p<0.001), while pioglitazone resulted in a significantly greater reduction in HOMA-IR (p=0.048), in line with findings from previous studies.³ No significant differences were observed between the groups for other parameters.

Adverse effects included lichen planus in one patient on metformin, nausea in four patients receiving metformin, and weight gain associated with pioglitazone.

In summary, our study indicates that both metformin and pioglitazone are effective in reducing the clinical and biochemical severity of AN. Metformin is preferable in overweight individuals due to its favourable effects on weight and BMI, while pioglitazone may be more beneficial for non-obese patients where higher insulin resistance predominates. Moreover, pioglitazone presents a viable alternative for those intolerant to metformin.

The strength of our study lies in its direct comparison of two commonly used insulin sensitisers on AN, evaluating both dermatologic and metabolic outcomes. However, limitations include the short 12-week duration and lack of dietary control, which may impact long-term assessments of efficacy. Lack of blinding and allocation concealment were the other limitations.