Sarcoidosis-like reactions in cancer patients treated with immune checkpoint inhibitors: experience in a Spanish hospital

Background Immune checkpoint inhibitors (ICI) have been associated with several immune-related adverse events, including sarcoidosis-like reactions (SLR). SLR, which has a low prevalence but an increasing incidence, is similar to sarcoidosis in terms of histology, and clinical and radiological manifestations. The most commonly affected organs are hilar and mediastinal lymph nodes and skin. SLR is an exclusion diagnosis, so a lymph node biopsy can be useful to distinguish between tumor progression and SLR, particularly in tumors in which nodal involvement is very common. Patients and methods We performed a retrospective analysis of SLR in all cancer patients receiving ICIs in our institution between January 2016 and June 2020. Results Among the 1063 treated patients, seven experienced SLR, four of whom were symptomatic (cough, skin lesions, arthralgia), with time to onset ranging from 1.5 to 6.7 months after ICI initiation. All seven patients had bilateral hilar lymphadenopathy, and granulomatous reactions in five of the six patients with lymph node biopsies. SLR improved in all patients, including four patients who continued with ICI. Three patients received corticosteroids and/or stopped ICI therapy. Four of these patients had partial responses at the time SLR was identified. Conclusion Management of SLR lacks a consensus recommendation, although corticosteroids and/or stopping the ICI are generally implemented. The potential consequences of stopping anticancer treatment should be taken into consideration, particularly in the absence of clear management recommendations (AU)

Sarcoidosis-like reactions in cancer patients treated with immune checkpoint inhibitors: experience in a Spanish hospital.

BACKGROUND: Immune checkpoint inhibitors (ICI) have been associated with several immune-related adverse events, including sarcoidosis-like reactions (SLR). SLR, which has a low prevalence but an increasing incidence, is similar to sarcoidosis in terms of histology, and clinical and radiological manifestations. The most commonly affected organs are hilar and mediastinal lymph nodes and skin. SLR is an exclusion diagnosis, so a lymph node biopsy can be useful to distinguish between tumor progression and SLR, particularly in tumors in which nodal involvement is very common. PATIENTS AND METHODS: We performed a retrospective analysis of SLR in all cancer patients receiving ICIs in our institution between January 2016 and June 2020. RESULTS: Among the 1063 treated patients, seven experienced SLR, four of whom were symptomatic (cough, skin lesions, arthralgia), with time to onset ranging from 1.5 to 6.7 months after ICI initiation. All seven patients had bilateral hilar lymphadenopathy, and granulomatous reactions in five of the six patients with lymph node biopsies. SLR improved in all patients, including four patients who continued with ICI. Three patients received corticosteroids and/or stopped ICI therapy. Four of these patients had partial responses at the time SLR was identified. CONCLUSION: Management of SLR lacks a consensus recommendation, although corticosteroids and/or stopping the ICI are generally implemented. The potential consequences of stopping anticancer treatment should be taken into consideration, particularly in the absence of clear management recommendations.

Clinical utility of plasma-based digital next-generation sequencing in patients with advance-stage lung adenocarcinomas with insufficient tumor samples for tissue genotyping.

BACKGROUND: Approximately 30% of tumor biopsies from patients with advanced-stage lung adenocarcinomas yield insufficient tissue for successful molecular subtyping. We have analyzed the clinical utility of next-generation sequencing (NGS) of cell-free circulating tumor DNA (ctDNA) in patients with inadequate tumor samples for tissue genotyping. PATIENTS AND METHODS: We conducted the study in a multi-institutional prospective cohort of clinically unselected patients with advanced-stage lung adenocarcinomas with insufficient tissue for EGFR, ALK or ROS1 genotyping across 12 Spanish institutions (n = 93). ctDNA NGS was carried out by Guardant Health (Guardant360, Redwood City, CA), using a hybrid-capture-based 73-gene panel. Variants were deemed actionable if they were part of the OncoKB precision oncology knowledge database and classified in four levels of actionability based on their clinical or preclinical evidence for drug response. RESULTS: Eighty-three out of 93 patients (89%) had detectable levels of ctDNA. Potentially actionable level 1-4 genomic alterations were detected in 53 cases (57%), of which 13 (14%) had level 1-2A alterations (Food and Drug Administration-approved and standard-care biomarkers according to lung cancer guidelines). Frequencies of each genomic alteration in ctDNA were consistent with those observed in unselected pulmonary adenocarcinomas. The majority of the patients (62%), particularly those with actionable alterations (87%), had more than one pathogenic variant in ctDNA. The median turnaround time to genomic results was 13 days. Twelve patients (13%) received genotype-matched therapies based on ctDNA results, deriving the expected clinical benefit. Patients with co-occurring pathogenic alterations had a significantly shorter median overall survival as compared with patients without co-occurring pathogenic alteration (multivariate hazard ratio = 5.35, P = 0.01). CONCLUSION: Digital NGS of ctDNA in lung cancers with insufficient tumor samples for tissue sequencing detects actionable variants that frequently co-occur with other potentially clinically relevant genomic alterations, allowing timely initiation of genotype-matched therapies.