NEW YORK – Patients with cancers of unknown primary (CUP) had more therapy options and better outcomes when their treatment was guided by a 90-gene expression assay developed by Canhelp Genomics compared to patients on standard chemotherapy, researchers reported at the European Society for Medical Oncology Congress on Sunday.
Cancers of unknown primary, also known as occult primary cancer, are histologically confirmed metastatic tumors, for which the primary site cannot be identified. CUP accounts for an estimated 2 percent to 9 percent of tumors. Patients with CUP tend to have poor prognoses, with median survival ranging from six months to nine months. Canhelp Genomics is betting that by identifying the site of origin of patients' cancers, they'll have the chance to receive more effective targeted treatments that improve their odds of survival.
The Canhelp-Origin test, which includes a 90-gene expression PCR assay and gene analysis software, is approved by China's National Medical Products Administration and is designed to distinguish 21 tumor types that comprise 95 percent of all solid tumors. In a multicenter validation study using 1,417 samples, Canhelp Genomics reported an overall accuracy of 94.4 percent for its test and a specificity greater than 99 percent. In a number of other pan-cancer studies and in difficult-to-diagnose tumor types such as metastatic brain and liver cancers, the test has shown more than 90 percent diagnostic accuracy.
While the test has proven ability to classify the tumor of origin across cancer types, researchers from Fudan University Shanghai Cancer Center sought to prove that treatment guided by Canhelp-Origin was superior to empiric chemotherapy in CUP patients. In a Phase III trial, which ran for enrolled patients between September 2017 and March 2021, researchers randomized 182 patients to either receive site-specific cancer therapy guided by the Canhelp-Origin test or empiric chemotherapy. In the molecularly guided treatment arm, most patients were matched to a targeted therapy specific to the cancer site of origin, while 26.4 percent received a standard chemotherapy regimen.
At a median follow-up of 42.9 months, the site-specific therapy group had a median progression-free survival of 9.6 months versus 6.6 months in the control group, a statistically significant three-month improvement. Median overall survival was 28.2 months in the site-specific treatment group versus 19.0 months in the control arm, but the 9.2-month improvement with molecularly informed treatment wasn't statistically significant.
Canhelp Genomics CEO Qinghua Xu said in an interview that the results from this study were "quite significant and practice changing." Xu added that guidelines from the China Anti-Cancer Association for CUP published in May includes updated recommendations on using tumor tissue-of-origin gene detection kits and gene analysis software for patients with CUP.
"By presenting this study at the ESMO Congress, we also wish to provide valuable evidence for ESMO and National Comprehensive Cancer Network's CUP guidelines," Xu said.
The latest study is also part of the company's efforts to further validate Canhelp-Origin as a tool for guiding therapy in diverse patient groups and generate data that can facilitate regulatory approval in other countries, including in the US and Europe. However, like in China, the company will seek approval for Canhelp-Origin as a tissue-of-origin test in other markets and not specifically for guiding therapy.
Based on the US Food and Drug Administration's 2018 510(k) clearance of Genecast's Tissue of Origin test, Xu doesn't expect it will be necessary to show that treatments given based on Canhelp-Origin's tissue-or-origin determination yield an overall survival benefit. "The registration trial for approval will be more focused on assessing the diagnostic accuracy of the assay across various types of cancers rather than guiding therapy," Xu said.
In terms of competition, Xu said that while Genecast's Tissue of Origin test is "very effective," it only covers 14 types of cancer, whereas the Canhelp-Origin test has demonstrated it can identify 21 tumor types. Another test developed by Hologic subsidiary Biotheranostics is CancerTYPE ID, which uses real-time PCR to measure expression of 92 genes in the tumor and identify 50 types of cancer. The lab-developed test is marketed as a tool that doctors can use for definitively diagnosing and subtyping tumors of uncertain origin, but unlike Genecast's test, it doesn't have FDA clearance.
At the American Society of Clinical Oncology's annual meeting last year, researchers from the Mayo Clinic and Biotheranostics used CancerTYPE ID combined with next-generation sequencing to determine the tumor type in 92.5 percent of 3,168 diagnostically ambiguous cases and characterize BRAF mutations in a subset of tumors. Based on this, patients with a variety of BRAF-mutated tumors were eligible for FDA-approved BRAF inhibitors or were able to receive targeted treatment within a clinical trial that they otherwise wouldn't have been able to access.
In another study at ESMO, researchers similarly investigated methods of facilitating CUP patients' access to targeted therapies by directly testing for actionable genetic alterations. In results from the CUPISCO trial, also presented at ESMO, researchers used comprehensive genomic profiling to inform treatment for newly diagnosed CUP patients and reported better progression-free and overall survival for those on molecularly guided therapy compared to patients on standard chemotherapy.
Meanwhile, studies have tried to demonstrate the value of using just gene expression profiling in improving treatment for CUP patients, but there isn't definitive evidence showing that therapy guided by such an assay is superior to empiric chemotherapy, said Xu, citing two examples. In a 2012 trial involving 252 patients with CUP in the US, researchers used a 92-gene cancer classification assay to identify the tissue of origin and provide site-specific treatment to 194 patients. The median overall survival time was 12.5 months compared to less than 10 months for historic control groups treated with empiric chemotherapy.
Xu said that trial was meaningful, but because it did not directly compare outcomes with a control arm, it did not establish superiority. A second trial, conducted by researchers from Japan, did not show an improvement in one-year survival for patients with CUP whose therapy was guided by gene expression profiling. "A major issue in this trial was that they did not use a fully validated gene expression assay, so there were some misclassified cases in the trial," said Xu.
In the Canhelp-Origin study, one major challenge, Xu said, was recruiting enough patients that truly had CUP. Guidelines require an extensive set of clinical diagnostic tests to identify the site of origin, including immunohistochemical tests, an in-depth medical history, a physical examination with blood tests, and CT scans of the chest, abdomen, and pelvis for all patients, mammography for women, and other symptom-specific tests. A CUP diagnosis is only confirmed if the diagnostic work-up does not definitively identify the primary tumor. "It requires a significant amount of time and work to recruit an adequate number of patients," Xu said.
The study took five years to complete due to the challenges of identifying eligible patients. Xu noted that Fudan University Shanghai Cancer Center, where Xichun Hu is director of medical oncology and principal investigator of the study, has established a specialized CUP center and represented an important recruitment site.
The effort researchers expended to find study participants ultimately paid off, according to Xu, as the trial showed that site-specific therapy guided by the Canhelp-Origin test gave patients additional treatment options they wouldn't have had otherwise. "This finding highlights the evolution of targeted therapies and immunotherapies for specific advanced tumor types, which can truly transform the medical landscape for CUP patients," Xu said.