Does the World Health Organization histological classification predict outcomes after thymomectomy? Results of a multicentre study on 750 patients.

Abstract

The World Health Organization (WHO) thymoma histological classification clinical value remains a controversy. In this study, we evaluated its prognostic significance in patients with thymoma treated with radical intent. Six high-volume Italian Thoracic Surgery Institutions collaborated with their own retrospective anonymized datasets. Demographic, clinical, pathological and treatment data were examined. A WHO histological classification (WHO-HC) collapsed scheme (A/AB and B1/B2 types merged) was proposed and compared with the traditional one. Predictors of survival were assessed using a Cox model with shared frailty. Competing-risk regression models were performed to identify the association between individual factors and freedom from recurrence. Between 1990 and 2011, 750 thymomas were operated on in participating centres. Myasthenia gravis was observed in 363 (48%) patients. A complete resection was achieved in 676 (91%) cases. One hundred and nine patients (15%) had a WHO-HC A type, 166 (22%) AB, 179 (24%) B1, 158 (21%) B2 and 135 (18%) B3. The rate of 5-year OS and cumulative incidence of recurrence for all cases was 91% and 0.11, respectively. Five-year survival rates by WHO-HC in the collapsed scheme were A/AB 93%, early-B 90% and advanced-B 85%. Masaoka stage only was demonstrated to be an independent predictor for survival and recurrence. The WHO-collapsed scheme showed a trend in influencing recurrence overall survival development (hazard ratio: 1.32; P = 0.16). Our results show evidence of lack of significance by WHO-HC in influencing prognosis, even though the proposed collapsed scheme revealed a fair stratification of risk to relapses and better correlation with patients' clinical characteristics.