Independent investigators reported today full results from the NEO3-09 study, reaffirming earlier top-line results that showed Lymphoseek® (99mTc-tilmanocept) met all primary and secondary endpoints and exhibited superior performance to vital blue dye in intraoperative lymphatic mapping (ILM) procedures. The results were presented during a moderated poster-discussion session today by Anne Wallace, MD, Moores Cancer Center, University of California San Diego, and Vernon Sondak, MD, H. Lee Moffitt Cancer Center in Tampa, Florida, at the American Society of Clinical Oncology (ASCO) Meeting in Chicago.
"NEO3-09 is intended to be the final Phase 3 study for initial U.S. registration of the lymphatic tissue tracing agent originally developed at UC San Diego Moores Cancer Center, where we went from bench to bedside to use this agent in helping diagnose breast cancer patients in clinical trials," said Dr. Wallace. "Lymphoseek compared very favorably to the standard of care, vital blue dye, was well tolerated without side effects, and as noted by prior studies, continued to show superior performance characteristics. As a surgeon who has worked with Lymphoseek for years, I find it easy to localize the sentinel node and appreciate its ease of use."
NEO3-09 Primary Endpoint Met - Strong Findings Across Both Phase 3 Studies
The primary endpoint of the NEO3-09 study was the comparison (the Concordance Rate, or the rate of agreement) of Lymphoseek versus vital blue dye, where vital blue dye is considered by FDA as the only approved, on-label agent for lymphatic mapping, thus making it the appropriate requisite "Truth Standard" comparator for registration purposes. The Concordance Rate was analyzed on both a per-node and per-patient basis.
In NEO3-09, study subjects yielded a total of 229 lymph nodes stained with vital blue dye. Of these blue-stained nodes, Lymphoseek detected 229, for a Concordance Rate of 100%, which was a highly statistically significant finding (p<0.0001). This Concordance Rate was consistent with the 97.67% rate observed in the NEO3-05 study (p<0.0001).
On a per-patient basis, the NEO3-09 study yielded a total of 133 patients with lymph nodes stained with vital blue dye. Of these patients, Lymphoseek detected the same blue-stained nodes in all 133 patients, for a Concordance Rate of 100%, a highly statistically significant finding (p<0.0001). This Concordance Rate was consistent with the 96.32% rate observed in the NEO3-05 study (p<0.0001).
Key Findings in Detection of Lymph Nodes Bearing Cancer
The NEO3-09 Phase 3 clinical study enrolled 133 subjects with either breast cancer or melanoma (n = 68 and 65 patients, respectively) in the intent-to-treat (ITT) population. On a per node basis, Lymphoseek exhibited a failed detection rate (FDR) of 0%, whereas vital blue exhibited an FDR of 25%. The prospective analysis confirmed the earlier retrospective analysis of Lymphoseek's lower FDR observed in the first Phase 3 study, NEO3-05. This low failed detection rate of tumor-bearing lymph nodes for Lymphoseek compared to vital blue dye means Lymphoseek missed fewer lymph nodes that contained cancer, a key finding given that the objective of ILM is to determine if cancer has spread to the lymph nodes.
Lymphoseek also exhibited a lower the FDR on a per-patient basis. Across the replicate NEO3-05 and NEO3-09 Phase 3 studies, among the 55 patients identified to have lymph nodes containing pathology-confirmed tumor, Lymphoseek missed 0 patients, for an FDR of 0%, whereas vital blue dye missed 4 patients (2 breast cancer and 2 melanoma diagnoses) for an FDR of 7.3%(p<0.044). Additionally, Lymphoseek also identified 2 patients with lymphoma that were not identified by vital blue dye. Thus, Lymphoseek facilitated the identification of 6 patients out of a total of 55 patients with lymph node-positive pathology (10.9%) whose cancer status would not have been accurately identified by vital blue dye.
"Both on a per-node and per-patient basis, the NEO3-09 data presented here at ASCO continue to reinforce the findings of previous clinical studies that suggest Lymphoseek provides significantly higher specificity and sensitivity, and a significantly lower failed detection rate, than vital blue dye," said Dr. Vernon Sondak of the H. Lee Moffitt Cancer Center in Tampa, Florida. "With the diagnostic and safety limitations of vital blue dye, these results should be a welcome sign for clinicians searching for potential innovative technologies to help in the diagnostic process for breast cancer or melanoma patients."
NEO3-09 Secondary Endpoint Findings
A secondary analysis treated Lymphoseek as the "Truth Standard" in ILM procedures; this Reverse Concordance Rate was also analyzed on a per-node and per-patient basis. In NEO3-09, 378 lymph nodes labeled with Lymphoseek were obtained. Of these, vital blue dye was observed in 229 nodes. Using Lymphoseek as the Truth Standard, the Reverse Concordance Rate for vital blue dye was 60.58%, which was not statistically significant.
In the per-patient analysis, the NEO3-09 study yielded a total of 152 patients with lymph nodes labeled with Lymphoseek. Of these patients, vital blue dye detected the same blue-stained nodes in only 76 patients, a Reverse Concordance Rate of 50.00%.
Using the Concordance Rate and Reverse Concordance Rate data from NEO3-09 in a pre-specified, prospective statistical test of superiority, Lymphoseek's performance was significantly superior to vital blue dye in lymph node detection (p<0.0001).
NEO3-09 Safety & LN Identification Findings
In both NEO3-09 and NEO3-05, Lymphoseek demonstrated no drug-related serious adverse events or clinically significant adverse events, whereas vital blue dye exhibited 3 adverse events, including one significant drug-related adverse event, anaphylactic hypotension. In over 500 subjects receiving Lymphoseek to date, no clinically significant drug-related adverse events have been reported.
In a full regional nodal dissection procedure, a patient with breast cancer or melanoma may have as many as 20 to 30 lymph nodes removed in order to determine whether or not cancer has spread to other parts of their body. The very invasive nature of such an extensive surgical procedure frequently causes significant side effects or morbidity (e.g., bleeding, pain, infection, neuropathy, seromas, and lymphedema). In the NEO3-05 and NEO3-09 studies combined, Lymphoseek detected an average of 2.4 lymph nodes per patient, whereas vital blue dye detected an average of approximately 1.5 lymph nodes per patient. With this small difference, Lymphoseek exhibited superior performance in detecting lymph nodes containing cancer, as evidenced by its lower FDR, noted above. The average number of lymph nodes detected by Lymphoseek is still far below the number of lymph nodes removed in a full nodal dissection procedure, thus potentially sparing the patient the morbidity and side effects associated with more complete regional nodal dissection procedures. In addition, in over twelve months of post-surgical follow-up to date of the patients involved in the NEO3-05 study, no morbidity issues with Lymphoseek have been reported to date.
"The results presented today reinforce prior clinical evidence indicating that Lymphoseek provides an improved modality for surgical oncologists to identify and diagnose potentially cancerous lymph nodes in breast cancer and melanoma versus the only currently approved agent, vital blue dye," said Dr. Fred Cope, Neoprobe's Senior Vice President, Pharmaceutical Research and Clinical Development. "Interestingly, these data also showed the potential ability for Lymphoseek to provide additional benefit to some patients by identifying cancerous lymph nodes that were missed by vital blue dye, which results in more accurate disease staging and ultimately patient treatment."
University of California San Diego