Cancer risk in East Asian patients associated with acquired haemolytic anaemia: a nationwide population-based cohort study

Victor C. Kok, Fung-Chang Sung, Chia-Hung Kao, Che-Chen Lin, Chun-Hung Tseng

This study investigated whether patients with acquired haemolytic anaemia (AHA) would have elevated cancer risk including that for non-haematological solid tumours. We further examined whether the cancer risk would be different between patients with autoimmune type AHA (AIHA) and patients of non-AIHA. Using nationwide population-based insurance claims data of Taiwan we identified a cohort of patients with AHA with no pre-existing cancer, (n = 3902) and a comparison cohort (n = 39020) without AHA, frequency-matched by gender, age, urbanization of residency and diagnosis date. Incidence and Cox method estimated adjusted hazard ratios (aHR) of cancers controlling covariates by the end of 2010 were calculated. Risks between patients with AIHA and non-AIHA were compared. Sensitivity analysis was carried out to measure the risk of cancer between patients with and without AHA by follow-up years. Patients with AHA had a 90 % greater incidence of cancer than controls, with an aHR of 1.78 (95 % confidence interval (CI), 1.50-2.12)]. The overall aHRs of cancer for patients with AIHA and non-AIHA were 2.01 (95 % CI, 1.56-2.59) and 1.87 (95 % CI, 1.53-2.29), respectively, compared with the comparison cohort. The aHRs for lymphatic-haematopoietic malignancy were 19.5 and 9.59 in the AIHA and non-AIHA cohorts, respectively. No hazard of colorectal, lung, liver or breast cancer was significant. There is a near 2-fold elevated risk for subsequent cancer in patients with AHA, particularly for lymphatic-haematopoietic malignancy, which is much greater for patients with AIHA than non-AIHA. These findings can help clinicians decide patient-centred personalized long-term management.