Growth rate and site of pulmonary metastasis to predict lung relapse and overall survival in patients affected by bone and soft tissue sarcomas (B-STS).

Abstract

11571 Background: Despite surgically resectable pulmonary metastases may lead to cure patients with B-STS (Chudgar NP 2017), a substantial proportion of patients will eventually relapse. Presently, patient selection is based on unique organ involvement, number of metastases, interval between previous surgery and pulmonary progression or relapse. We assessed the impact of anatomical site of metastasis into the lung (as if the pleural site might ease further tumor spreading) and nodule growth rate as additional predictive/prognostic factors of lung progression-free survival (L-PFS) and overall survival (OS). Methods: In our prospectively collected database, we retrospectively evaluated patients operated for B-STS pulmonary progression at 3 different centers from 2005 to 2019. Beyond patients’ clinical features at both baseline and disease progression in the lungs, we focused on whether the relapse occurred into the parenchyma or nearby the pleura (Welter S 2012); secondly, we estimated lung metastasis growth rate, defined as tumor doubling time (TDT) (Nakamura T 2011). Statistical analyses were carried out with IBM SPSS (v. 20.0). Survival outcomes were estimated by Kaplan-Meier method. Hazard ratios (HR) were estimated by Cox regression. Multivariate analysis was performed for both L-PFS and OS according to Cox proportional hazard model. All tests were 2-sided with their corresponding 95% confidence intervals (CI95%). Results: We identified 138 patients who underwent lung metastasectomy [(F=66 (48%); median age at surgery 50 (14-78)]. Median PFS and L-PFS were 8.7 months (CI95% 6.6-10.9) and 8.6 months (CI95% 6.2-11.0), respectively. Median OS was 40.6 months (CI 95% 32.8-48.5). Univariate analysis showed a statistically significant impact of the following variables for both L-PFS and OS: ECOG 0, nodule number <3, being disease-free after first-line treatment, no pleural involvement, and TDT >40 days. Disease-free interval ≤ 24 months and absence of metastases at diagnosis showed significant correlation with L-PFS and OS, respectively. At multivariate analyses the following variables retained statistical significance for L-PFS: TDT >40 days (HR 0.53, CI95% 0.31-0.93, p=0.028); nodule number <3 (HR 0.54, 95%CI 0.29-0.99, p=0.048), no pleural involvement (HR 0.39, CI95% 0.22-0.70, p=0.001); and for OS: TDT >40 days (HR 0.36, CI95% 0.18-0.72, p=0.004), nodule number <3 (HR 0.35, 95%CI 0.18-0.71, p=0.004), no pleural involvement (HR 0.49, CI95% 0.24-0.98, p=0.045), and ECOG 0 (HR 0.29, 95%CI 0.14-0.59, p=0.001). Conclusions: Acknowledging its retrospective nature and the need for an external validation, our series highlights the key-role of the anatomical site of relapse within the lung and the impact of tumor growth rate. If confirmed, these two clinical parameters should be factored in the decision making on performing pulmonary metastasectomy.