LAUSR

BackgroundHistological subdivision into typical (TC) and atypical (AC) is crucial for treatment and prognosis of lung carcinoids but can be also very challenging, even for experts. In this study, we aimed to strengthen or reduce the prognostic value of several pathological, clinical, or per-operative factors some of which are still controversial.MethodsWe retrospectively reviewed clinical records related to 195 patients affected by TC (159) or AC (36) surgically treated between 2000 and 2014, in three different centers. Survival and subtypes comparison analyses were performed to identify potential prognostic factors.ResultsTCs showed a lower rate of nodal involvement than ACs (N0 = 94.9%; N1 = 1.9%; N2 = 3.2% in typical and N0 = 63.8%; N1 = 16.6%; N2 = 19.4% in atypical carcinoids, respectively, p < 0.0001). Long-term oncological results of resected carcinoids were significantly better in TCs than ACs with higher 5- and 10-year overall survival rates (97.2 and 88.2% vs. 77.9 and 68.2%, respectively; p = 0.001) and disease-free survival rates (98.2 and 90.3% in typical and 80.8 and 70.7% atypical carcinoids, respectively; p = 0.001). Risk factors analysis revealed that AC subtype [HR 4.33 (95% CI 1.72–8.03), p = 0.002], pathological nodal involvement [HR 3.05 (95% CI 1.77–5.26), p < 0.0001], and higher SUVmax [HR 4.33 (95% CI 1.03–7.18), p = 0.002] were independently and pejoratively associated with overall survival. Factors associated with a higher risk of recurrence were AC subtype [HR 6.13 (95% CI 1.13–18.86), p = 0.002]; nodal involvement [HR 5.48 (95% CI 2.85–10.51), p < 0.0001]; higher Ki67 expression level [HR 1.09 (95% CI 1.01–1.20), p = 0.047]; and SUVmax [HR 1.83 (95% CI 1.04–3.23), p = 0.035].ConclusionSurgery for lung carcinoids allows satisfactory oncological results which mainly depend on carcinoid subtype dichotomy, pathological nodal status, and SUVmax.