The FDA has granted priority review to a new drug application seeking the approval of the optical imaging agent Lumisight for use in patients with breast cancer, according to an announcement from Lumicell, Inc.
The FDA has granted priority review to a new drug application seeking the approval of the optical imaging agent Lumisight (pegulicianine) for use in patients with breast cancer, according to an announcement from Lumicell, Inc.1 The agency also accepted a premarket approval application for the Lumicell Direct Visualization System (DVS).
The DVS was designed to leverage Lumisight to identify, view, and guide the resection of residual cancer during the initial lumpectomy.1,2 The imaging agent is administered preoperatively, on the same day of the procedure, to highlight cancerous cells. Then, a hand-held imaging probe scans inside the breast cavity to identify activated Lumisight in residual cancer. Patient-calibrated tumor detection software provides real-time images from the cavity that can guide the surgeon in removing remaining disease.
The applications were based on findings from over 700 patients with breast cancer who had been enrolled across 5 studies conducted at academic and community cancer centers. Specifically, the safety and efficacy of the system and imaging agent was evaluated as part of the phase 3 INSITE trial (NCT03686215) and a phase 2 feasibility study (NCT03321929).
Data from the prospective, multicenter INSITE trial were recently published in NEJM Evidence.3 The trial enrolled female patients who were undergoing lumpectomy for stage I to III invasive breast cancer and/or ductal carcinoma in situ (DCIS). These patients were at least 18 years of age and received standard preoperative isotope injections, with gamma-probe scanning prior to surgery to confirm sentinel node identification by isotope. If they were receiving neoadjuvant therapy or were undergoing margin re-excision after previous lumpectomy, they were not included.
Two to 6 hours prior to undergoing surgery, study participants were given 1.0 mg/kg of Lumisight via intravenous infusion. Standard lumpectomy was then completed, and all specimens were oriented with sutures and/or six-color inking in the operating room to enhance orientation. Notably, patients were randomly assigned 10:1 to receive Lumisight-guided surgery or a control group that did not receive the imaging agent.
Allocation information was only revealed after the surgery was completed. Due to this, randomization was not designed to provide a control group to evaluate performance of the device. Each patient who underwent Lumisight-guided surgery served as their own control, with evaluation based on final margin pathology after standard surgery combined with additional imaging-guided cavity margins. Those in the control group were only included in the safety analysis.
Three primary end points were evaluated for the trial: percentage of patients for whom Lumisight-guided margins contained residual cancer following lumpectomy (success defined as greater than 3% for lower bound of the 95% CI), the percentage of margins with tumor that were Lumisight positive (sensitivity), and the percentage of margins with no tumor that were Lumisight negative (specificity). Secondary end points comprised positive margin rate following removal of Lumisight-guided margins and impact of Lumisight-guided margins on the volume of excised tissue.
Of the 406 patients enrolled to the trial, 14 withdrew prior to randomization. Among the remaining 392 patients, the median age was 64 years (range, 36-83). Moreover, 19.4% had DCIS only; 69.9% had invasive ductal carcinoma with or without DCIS, 9.9% had invasive lobular carcinoma with or without DCIS, and 0.8% had invasive lobular carcinoma with DCIS.
Following lumpectomy, patients were randomly assigned to the Lumisight-guided group (n = 357) or to the control group (n = 35).
Data indicated that of those in the investigative group, Lumisight-guided margins removed residual cancer following the procedure in 7.6% (n = 27), which was greater than 3%, meeting the criteria for success. Of these patients, 19 had all negative margins on standard pathology evaluation. Investigators noted that without the intervention, the residual disease would not have been detected. In 22 patients, the imaging agent served to remove residual cancer from margins that were negative after surgery on standard evaluation.
Additional findings showed that margin-level specificity was higher than the prespecified lower-bound goal of more than 60%, at 85.2% (95% CI, 83.7%-86.6%). Across all orientations, margin-level sensitivity was 49.3% (95% CI, 37.0%-61.6%). Investigators noted that the trial did not meet the prespecified lower-bound goal of greater than 40% for sensitivity. Data from a post-hoc analysis indicated what when calculating sensitivity by just examining orientations where histopathology was available for comparison, the sensitivity rate was higher, at 58.6% (95% CI, 44.9%-71.4%).
Across all margins, the margin-level negative predictive value of the imaging intervention was 98% (95% CI, 97.7%-98.8%), and the margin-level positive predictive value was 9.2% (95% CI, 6.4%-12.6%).
Additionally, 17.4% (n = 62) of patients had at least 1 positive margin following surgery and prior to the imaging agent. Notably, in 14.5% of these patients, removal of Lumisight-guided shaves led to a conversion of all positive margins to negative margins during initial surgery, which prevented the need for a second surgical procedure.
Previously, in May 2022, findings from the feasibility study were published in JAMA Surgery.4 The nonrandomized, controlled trial included female patients who had invasive breast cancer and/or ductal carcinoma in situ and were at least 18 years of age. Here, participants were given Lumisight at the same dose, 1 mg/kg, and went on to receive BCS, which was defined as resection of the main tumor specimen and any shaved cavity margins required to obtain grossly negative margins by surgeon assessment.
A total of 234 patients were enrolled to the study. All patients received the imaging agent, but 4 withdrew before completing the procedure; the remaining 230 patients were analyzed for performance metrics.
The association of the imaging system with shave margin pathology was assessed by using the truth standard hierarchy approach. Of 1091 negative images, 1072 were noted to be true negatives, which equated to a 98% negative predictive value. Imaging was positive in 43 of the 62 total positive margins, which equated to a sensitivity rate of 69.4%. The false-negative rate of the imaging procedure was 1.2%. The per-margin sensitivity rate with Lumisight was 69.4% vs 38.2% compared with routine pathology assessment of lumpectomy specimens.
The association of the imaging intervention and final margin status was also evaluated. Of the 230 total patients, 83.5% achieved negative margins after BCS. In this group, 115 patients had additional image-guided excisions and 14 were found to have residual disease in the additionally excised margins; 11 patients had negative margins at the time of the final pathology assessment.
Moreover, 16.5% of the 230 patients were found to have positive margins after BCS. In this subset, 29 had positive imaging in at least 1 orientation; 23 of these patients had additional image-guided excisions, and 12 of which had residual disease detected in the final margins. Sixty percent of the 230 patients had additional excisions guided by the imaging system and 19% had residual disease. The false-negative rate of the Lumisight intervention was 23.7% on a per-patient level and the sensitivity rate at this level was 76.3%.
Previously, the FDA granted a fast track designation to Lumisight and a breakthrough therapy device designation to the Lumicell DVS.1