Brief Report: Chylothorax and Chylous Ascites During RET Tyrosine Kinase Inhibitor Therapy

Abstract

Introduction: Spontaneous chylous effusions are rare; however, they have been observed by independent investigators in patients treated with RET tyrosine kinase inhibitors (TKIs).

Methods: This multicenter, retrospective study evaluated the frequency of chylous effusions in patients treated with RET TKIs. Clinicopathologic features and management of patients with chylous effusions were evaluated.

Results: A pan-cancer cohort of 7517 patients treated with one or more multikinase inhibitor or selective RET TKI and a selective TKI cohort of 96 patients treated with selpercatinib or pralsetinib were analyzed. Chylous effusions were most common with selpercatinib (7%), followed by agerafenib (4%), cabozantinib (0.3%), and lenvatinib (0.02%); none were observed with pralsetinib. Overall, 12 patients had chylothorax, five had chylous ascites, and five had both. Time from TKI initiation to diagnosis ranged from 0.5 to 50 months. Median fluid triglyceride level was lower in chylothoraces than in chylous ascites (397 mg/dL [interquartile range: 304-4000] versus 3786 mg/dL [interquartile range: 842-6596], p = 0.035). Malignant cells were present in 13% (3 of 22) of effusions. Chyle leak was not identified by lymphangiography. After initial drainage, 76% of patients with chylothorax and 80% with chylous ascites required additional interventions. Selpercatinib dose reduction and discontinuation rates in those with chylous effusions were 47% and 0%, respectively. Median time from diagnosis to disease progression was not reached (95% confidence interval: 14.5-undefined); median time from diagnosis to TKI discontinuation was 11.4 months (95% confidence interval: 8.2-14.9).

Conclusions: Chylous effusions can emerge during treatment with selected RET TKIs. Recognition of this side effect is key to prevent potential misattribution of worsening effusions to progressive malignancy.

Keywords: Chylothorax; Chylous ascites; Non–small cell lung cancer; RET tyrosine kinase inhibitor; Selpercatinib; Thyroid cancer.

Figure 1.. Chylous Effusion Acquisition and Characteristics.

(A) A competing risk analysis was performed for chylous effusion acquisition in a sub-cohort of Cohort 1. Patients with and without a diagnosis of chylous effusion that had longitudinal follow-up data on RET TKI therapy were eligible (n=131). The event of interest was the occurrence of chylous effusion while the competing event was death or discontinuation of RET TKI therapy. There were no censored patients as all patients in the cohort experienced at least one event. The table shows the percentage of patients who experienced either a chylous event (red) or death/TKI discontinuation (blue) at one, two, and three years from the initiation of RET TKI therapy. (B) Grossly white/cream-like pleural fluid obtained from a 70-year-old man with KIF5B-RET fusion-positive lung adenocarcinoma while on selpercatinib. The patient elected to undergo serial thoracentesis and declined pleurodesis or tunneled pleural catheter placement. (C) Yellow pleural fluid obtained from a 66-year-old man with KIF5B-RET fusion-positive lung adenocarcinoma treated with selpercatinib. The patient underwent prior thoracentesis for a non-chylous effusion before initiation of selpercatinib; however, his pleural effusion recurred requiring repeat thoracentesis which revealed a chylous pleural effusion on selpercatinib. The patient ultimately underwent tunneled pleural catheter placement. (D) Orange pleural fluid obtained from a 66-year-old woman with renal cell carcinoma while on treatment with lenvatinib. She underwent tunneled pleural catheter placement with subsequent rapid pleurodesis and removal of the tunneled pleural catheter 24 days later. Pleural fluid triglyceride levels are shown for panels A-C. (E) The frequency of radiographic laterality in patients with a biochemically confirmed chylothorax across the study cohort (N=17). (F) Pleural fluid triglyceride levels in patients treated with selpercatinib (N=10) vs. multikinase inhibitors (N=6). (G) Ascitic fluid triglyceride levels in patients treated with selpercatinib (N=7) vs. multikinase inhibitors (N=3) (small numbers limit further statistical analysis). (H) Triglyceride levels in pleural (N=16) vs ascitic (N=10) fluid across the study cohort.

Figure 2.. Patient Course and Management.

Clinical course of the full study cohort (N=22). The start time represents the initiation of the first RET tyrosine kinase inhibitor patients received. MGH; Massachusetts General Hospital; MSK, Memorial Sloan Kettering Cancer Center; UC, University of Chicago Hospital.