Thrombocytosis in primary lung cancer

A B S T R A C T BACKGROUND The prevalence of thrombocytosis in lung cancer has been variably re- ported from 13% to 60%. The vast majority of published data derive from retrospec- tive studies and therefore the estimation of the real prevalence is difficult. PATIENTS AND METHODS We prospectively enumerated platelets in the peripheral blood of 317 consecutive, untreated lung cancer patients. Thrombocytosis was defined as a pletelet count >400.000/mm3. Tumours were histopathologically classified ac- cording to the WHO classification and staging was performed according the 2002 version ISS for NSCLC and to the VALSG two stage system for SCLC. RESULTS Thrombocytosis was present in 64 (20.2%) patients and it was unrelated to gender and extent of disease but it was age related: 29/91 (31.8%) in patients ≤50 yrs versus 35/226 (15%) in patients >50yrs old (p=0.001). The distribution of thrombo- cytosis in the two major histological types was 54/165 (32.7%) for NSCLC and 10/152 (6.6%) for SCLC (p<0.001). Thrombocytosis did not differ (p=0.13) between NSCLC subtypes: adenocarcinomas (37/95, 39%), squamous cell carcinomas (11/48, 23%) and large cell carcinomas (6 22, 27%). Time to progression (TTP) did not differ (p=0.2) between thrombocytemic and non-thrombocytemic patients. On the contrary, 142/253 (56%) non-thrombocytemic patients responded to chemotherapy versus 21 of 64 (32.8%) thrombocytemic patients (p=0.001). CONCLUSIONS Thrombocytosis is rather uncommon in untreated patients with SCLC while it is present in about one third of patients with NSCLC and it is unrelated to clinical stage or gender but related to age. In SCLC, it is unrelated to stage, gender or age. The presence of thrombocytosis in NSCLC and SCLC does not affect TTP. It is inversely related to response to chemotherapy but only in NSCLC.