Cancer risk in a large acromegaly cohort: a standardized incidence ratio analysis

To evaluate overall and site-specific cancer risk in a large Turkish acromegaly cohort and assess the contribution of surveillance bias.

We retrospectively analyzed 393 acromegaly patients followed at a single tertiary center (1980-2023). Standardized incidence ratios (SIRs) were calculated against Turkish national cancer rates. Sensitivity analyses included thyroid exclusion, graded landmarks (1, 2, 5 years), temporal restrictions, and reference rate variation (± 20%). Multivariable logistic regression assessed predictors of cancer, including cumulative IGF-1 exposure (time-weighted average, TWAvg×ULN).

Sixty-four patients (16.3%) developed cancer over a median follow-up of 9 years. The overall cancer SIR was 5.48 (95% CI: 4.15-7.10), driven by thyroid (46.25), renal/urinary (7.38), and breast cancer (3.75; all p < 0.001 except breast p = 0.006). The non-thyroid SIR remained significant at 2.76 (1.86-3.94, p < 0.001) and stable across graded landmarks (range 2.42-2.53; all p < 0.001) and other sensitivity analyses. TWAvg IGF-1 was higher in the cancer group (1.26 vs. 0.97 ×ULN; p = 0.002) and independently predicted cancer (OR: 1.80; 95% CI: 1.21-2.69; p = 0.004).

In this large single-center Turkish acromegaly cohort, the overall SIR (5.48) exceeded the pooled population-based estimate of 1.45. The non-thyroid SIR remained significant across all sensitivity analyses (range 2.22-3.45), indicating multi-site predisposition beyond surveillance bias. Cumulative IGF-1 exposure independently predicted cancer, supporting sustained hormonal excess as a driver of oncogenesis. These findings support systematic thyroid screening and heightened vigilance for non-thyroid malignancies in tertiary-managed acromegaly.