Preoperative predictors for recurrence sites associated with poor post-recurrence survival after surgery of non-small cell lung cancer: a multicenter study
Tetsuya Isaka,Hiroyuki Adachi,Kotaro Murakami,Jun Miura,Noritake Kikunishi,Naoko Shigeta,Yujin Kudo,Yoshihiro Miyata,Morihito Okada,Norihiko Ikeda,Hiroyuki Ito
DOI: https://doi.org/10.1186/s12885-023-11582-y
IF: 4.638
2023-11-06
BMC Cancer
Abstract:Background: The recurrence site that influences post-recurrence survival (PRS) in patients with non-small cell lung cancer (NSCLC) undergoing surgery and the preoperative predictors of recurrence remain unclear. Methods: Cohorts 1 and 2 had 4520 (who underwent complete resection for p-stage 0-IIIA NSCLC) and 727 (who experienced recurrence after surgery) patients, respectively. The initial sites of recurrence were the lungs (309 cases), thoracic lymph nodes (225 cases), pleura (112 cases), bone (110 cases), central nervous system (86 cases), adrenal gland (25 cases), abdomen (60 cases), cervical and axillary lymph nodes (38 cases), chest wall (13 cases), skin (5 cases), and eye and tongue (3 cases). For cohort 2 analysis, the initial recurrence site that resulted in poor PRS was analyzed by multivariable analysis using a Cox proportional hazard model. For cohort 1 analysis, the preoperative predictors of recurrence patterns with poor PRS were analyzed by multivariable analysis using a logistic regression model. Results: In cohort 2 analysis, recurrence in the central nervous system (hazard ratio [HR], 1.70; p < 0.001), bone (HR, 1.75; p < 0.001), abdomen (HR, 2.39; p < 0.001), and pleura (HR, 1.69; p < 0.001) were independent poor prognostic recurrent sites for PRS and they were high-risk sites (HRS). Intrathoracic lymph nodes, cervical and axillary lymph nodes, lungs, chest wall, adrenal gland, eye and tongue, and skin were low-risk sites (LRS) that did not affect PRS. Patients with multiple LRS without HRS recurrence had a worse prognosis than those with a single LRS without HRS recurrence (5-year PRS 20.2% vs. 37.7%, p < 0.001) and were comparable to those with HRS recurrence (p = 1.000). In cohort 1 analysis, preoperative predictors for HRS and multiple LRS recurrences were positron emission tomography (PET) maximum standardized uptake value (maxSUV) ≥ 3.2 (HR, 5.09; p < 0.001), clinical nodal metastasis (HR, 2.00; p < 0.001), tumor size ≥ 2.4 cm (HR, 1.96; p < 0.001) and carcinoembryonic antigen (CEA) ≥ 5 ng/ml (HR, 1.41; p = 0.004). The cumulative incidence rates of HRS and multiple LRS recurrences within 5 years were 55.9%, 40.9%, 26.3%, 11.1%, and 3.5% (p < 0.001) in patients with 4, 3, 2, 1 and 0 of the above risks, respectively. Conclusions: HRS and multiple LRS were vital recurrences associated with poor PRS. Preoperative PET maxSUV, clinical nodal metastasis, tumor size, and CEA level predicted the incidence of vital recurrence.