Assessing the early returns of robotic-assisted bronchoscopy in a patient population at high risk for cancer: factors associated with a diagnostic procedure and pathologic outcomes

Abstract

Abstract Background Lung cancer remains the most common cause of cancer-related deaths in the USA. Successful implementation of lung cancer screening programs has led to increased detection of peripheral pulmonary lesions (PPL). Robotic-assisted bronchoscopy (RB) is a relatively novel tool used to safely diagnose PPN. Additionally, with the developing precision of personalized medicine and targeted therapy, obtaining adequate tissue for next-generation sequencing (NGS) has become critical to optimizing the medical management of lung cancer. Our aim is to assess the diagnostic yield of RB, identify patient and procedure characteristics associated with performing a diagnostic biopsy, and evaluate the adequacy of tumor cellularity for biomarker identification. Results We performed a retrospective observational study consecutively enrolling 54 adult patients who underwent RB between January 2022 and March 2023. Records were reviewed for medical comorbidities and smoking status. PPL characteristics included size, location, presence of a bronchus sign on mapping computed tomography (CT) scans, distance from airway and pleura on CT, and days between obtaining CT and the bronchoscopy. Procedural factors included the radial endobronchial ultrasound (r-EBUS) view obtained and the sampling method(s) with which a diagnosis was achieved. Records were reviewed for final pathology and the need for further procedures to establish a diagnosis. Samples that yielded non-small cell lung cancer (NSCLC) were sent for an NGS panel if deemed adequate. The panels were reviewed to identify potentially actionable mutations. Multivariable logistic regression was performed to assess patients and lesions characteristics associated with a diagnostic biopsy. We sampled 56 lesions, 45 (80.4%) were from current or former smokers. The overall diagnostic yield was 73.2% and 68.9% for nodules less than 3 cm. Lesions size ranged from 8 mm nodules to 70 mm masses, with an average size 20.3 mm. Mean nodule size from diagnostic procedures was 21.5 mm and 16.9 mm from non-diagnostic ones (p = 0.04). Logistic regression analysis showed that the presence of a bronchus sign was associated with a diagnostic procedure (OR 19.4, p = 0.023), while lack of a r-EBUS view predicted a non-diagnostic bronchoscopy (OR 0.02, p = 0.013). 28 diagnostic biopsies (68.3%) yielded NSCLC and of the 25 cases that we followed, 22 samples (88%) were adequate for NGS. Conclusions The presence of a bronchus sign on CT and obtaining r-EBUS view during RB were associated with a higher diagnostic yield. RB successfully sampled PPL with adequate tumor cellularity for NGS.

Keywords

• Robotic bronchoscopy
• Diagnostic yield
• Peripheral pulmonary lesion
• Lung cancer
• Next-generation sequencing