Blood cholesterol and outcome of patients with cancer under regular cardiological surveillance

Abstract

Cardiovascular (CV) diseases and cancer share several similarities, including common risk factors. In the present investigation we assessed the relationship between cholesterol levels and mortality in a cardiooncological collective. In total, 551 patients receiving anticancer treatment were followed over a median of 41 (95% CI 40, 43) months and underwent regular cardiological surveillance. A total of 140 patients (25.4%) died during this period. Concomitant cardiac diseases were more common in patients who deceased (53 (37.9%) vs. 67 (16.3%), p < 0.0001), as well as prior stroke. There were no differences in the distribution of classical CV risk factors, such as hypertension, diabetes or nicotine consumption. While total cholesterol (mg/dL) was significantly lower in patients who deceased (157 ± 59 vs. 188 ± 53, p < 0.0001), both HDL and LDL cholesterol were not differing. In addition, cholesterol levels varied between different tumour entities; lowest levels were found in patients with tumours of the hepatopancreaticobiliary system (median 121 mg/dL), while patients with melanoma, cerebral tumours and breast cancer had rather high cholesterol levels (median > 190 mg/dL). Cholesterol levels were significantly lower in patients who died of cancer; lowest cholesterol levels were observed in patients who died of tumours with higher mitotic rate (mesenchymal tumours, cerebral tumours, breast cancer). Cox regression analysis revealed a significant mortality risk for patients with stem cell transplantation (HR 4.31) and metastasised tumour stages (HR 3.31), while cardiac risk factors were also associated with a worse outcome (known cardiac disease HR 1.58, prior stroke/TIA HR 1.73, total cholesterol HR 1.70), with the best discriminative performance found for total cholesterol (p = 0.002).