Diagnostic AI Tool for Identifying Early-Stage Lung Cancer Earns FDA 150(k) Clearance
The Virtual Nodule Clinic artificial intelligence–powered clinical decision support tool was cleared by the FDA for detection of early-stage lung cancer in lung nodules detected by a CT scan.
FDA 150(k) clearance was granted to the artificial intelligence (AI)–powered clinical decision support software Virtual Nodule Clinic for use by pulmonologists and radiologists managing patients with lung nodules that have the potential to be early-stage lung cancer, according to Optellum the company responsible for developing the device.1
This represents the first application of decision support guided by AI for use in early lung cancer diagnosis to receive FDA clearance.
“We are delighted to launch the world’s first AI-based decision support for early lung cancer diagnosis cleared by the FDA,” Václav Potěšil, PhD, co-founder and CEO of Optellum, said in a press release. “This clearance will ensure clinicians have the clinical decision support they need to diagnose and treat lung cancer at the earliest possible stage, harnessing the power of physicians and AI working together—to the benefit of patients. Our goal at Optellum is to redefine early diagnosis and treatment of lung cancer, and this FDA clearance is the first step on that journey. We look forward to empowering clinicians in every hospital, from our current customers at academic medical centers to local community hospitals, to offer patients with lung cancer and other deadly lung diseases the most optimal diagnosis and treatment.”
Use of Virtual Nodule Clinic has demonstrated ability to improve accuracy of diagnosis and support clinical decision making from examination of CT scan images. In a study of 300 cases of solid and semi-solid nodules, 150 each being benign or cancerous, statistically significantly improved accuracy of diagnoses were seen with the use of the tool in physicians across a range of expertise levels and specialties. Twelve readers were assessed, 6 each being pulmonologists or radiologists, with each performing 2 sequential reads of the cases, first blinded then unblinded to the AI score.2
The primary end point was the change in the area under the curve (AUC), the mean of each for blinded and unblinded rounds were 81.9% (95% CI, 80.5%-83.3%) versus 88.8% (95% CI, 87.7%-89.8%), respectively. This resulted in a mean effect size improvement of 6.85 AUC points (95% CI, 4.29%-9.41%; P < .001).
The study’s secondary end points were both improved. Overall, there was a 26% improvement in clinical recommendation with the use of AI and inter-reader consistency was reduced from 16.65 pp when blinded to 11.07 pp when assisted. When assessing sensitivity and specificity with a limit of magnitude threshold of 5%, results were improved when unblinded to AI at 97.89 and 42.28, respectively, versus 94.06 and 37.44 when blinded (P <.001). With a limit of magnitude threshold of 65%, corresponding measures were 65.00 and 89.28 for assisted versus 55.67 and 86.50 for unassisted (P < .001).
“This study demonstrates the clinical impact of the LCP [Lung Cancer Prediction] score,” Anil Vachani, MD, MS, Principal Investigator of the study and Associate Professor and Co-Director of the Lung Cancer Screening at the University of Pennsylvania, said in a press release. “When using the LCP score, all readers in the study significantly improved both their sensitivity and specificity of diagnosis, and readers at all levels of expertise became more consistent. This is significant because it could assist with early lung cancer diagnosis and intervention in today’s clinical practice, where many patients with cancerous nodules may face delays in diagnosis and treatment, while patients with benign nodules are often unnecessarily exposed to aggressive procedures with sometimes life-threatening complications.”
The LCP score is calculated using a full pattern of 3D pixels in standard images from CT scans, making it an attractive tool due to its availability for use in all health care settings.
References
1. Optellum Receives FDA Clearance for the World’s First AI-Powered Clinical Decision Support Software for Early Lung Cancer Diagnosis. New release. Optellum. March 23, 2021. Accessed March 24, 2021. https://www.businesswire.com/news/home/20210323005236/en
2. Vachani A, Massion PP, Munden RF, et al., Imaging AI Radiomics decision support improves physicians’ stratification of indeterminate pulmonary nodules: An MRMC study presented to the American Cancer Society National Lung Cancer Roundtable (NLCRT) 2020. Presented at: National Lung Cancer Roundtable. Accessed March 24, 2021. https://vimeo.com/487367357
Eric Ko on Combining Radiation Therapy With Immunotherapy to Treat Non–Small-Cell Lung Cancer
September 18th 2018ONCOLOGY spoke with Eric Ko, MD, PhD, who recently published a review article with his colleagues on strategies for combining radiation therapy with immunotherapy for the treatment of non–small-cell lung cancer.
