Skip to main content
Premium Trial:

Request an Annual Quote

Mt Sinai Study Uses NGS Panels to ID Personalized Biomarkers in Gynecologic Cancer Patients


NEW YORK (GenomeWeb) – Ovarian cancer is the leading cause of death for all gynecologic cancers with an estimated 14,000 deaths likely to occur this year, according to the American Cancer Society. The five-year survival rate is less than 50 percent, a number that has hardly budged in the last 40 years.

One reason for the dismal outlook is that while other cancers have benefitted from the discovery of biomarkers that can point to a potential therapy or predict whether a disease is likely to be more aggressive, there is only one such marker for ovarian cancer, and it is not very good.

Researchers at the Icahn School of Medicine at Mt. Sinai are hoping to change this, though, through a personalized cancer program that uses a combination of technologies to profile patients' tumors and design personalized noninvasive tests based on specific markers to monitor their response to treatment and recurrence.

A team led by John Martignetti has launched a study in which patients with ovarian or endometrial cancer who come to Mt. Sinai and are scheduled for surgical resection are eligible to enroll. The surgeon collects tumor and blood specimens, as well as ascites fluid, from patients that have consented. During patients' subsequent visits, the physician collects additional blood samples for testing.

The blood, tumor, and ascites fluid samples are used for subsequent research, including designing patient-derived cell lines on which targeted therapies can be tested, and all the information is stored in a HIPAA-compliant database.

One of the primary goals, though, is to develop a sequencing-based panel that is specific for gynecological cancers and that can be used to screen every patient and then design subsequent noninvasive tests to monitor the patient throughout treatment.

The team has been using a targeted sequencing strategy, including a hotspot panel and full-gene sequencing of TP53.  So far, they have sequenced about 40 patients, all of whom have had "multiple rearrangements," Martignetti said.

They use the results of the sequencing to then design a targeted droplet digital PCR-based test that can be used noninvasively. In their study, they are comparing the ability of the ctDNA test to detect the tumor markers and predict disease recurrence, relapse, or treatment resistance, to the US Food and Drug Administration-approved biomarker, CA125, as well as to known clinical outcomes, to see if the tumor-specific mutations present in circulating blood can serve as a more accurate diagnostic earlier in a patient's disease.

CA125 is not a good biomarker, but it is currently the only FDA-approved one for ovarian cancer. For about 20 percent of patients, CA125 levels never rise or fall, making it completely ineffective. Even in those patients where CA125 levels do change, its sensitivity and specificity for predicting recurrence range between 62 percent and 94 percent and 91 percent and 100 percent, respectively. And, "monitoring has never been shown to increase survival, and may in fact decrease quality of life," Martignetti said.

Hence, the search for a more universal, more predictive biomarker.

So far, Martignetti's team has sequenced around 40 patients with a hotpsot panel on Thermo Fisher's Ion Torrent PGM, and it has sequenced the entire TP53 gene on the Pacific Biosciences RS II instrument.

He described one patient who had a fusion between FGFR2 and FAM76A.  The fusion was tracked throughout the patient's disease course and compared with CA125 monitoring. At the time of surgery, CA125 was fairly elevated, Martignetti said. Between 2009 and 2013, the patient had 28 blood samples drawn, all by the same nurse. Half of the blood sample was used for testing CA125 and half to screen for the FGFR2/FAM76A fusion. During her disease course, CA125 was abnormal in only three out of the 28 times, despite the fact that she experienced recurrence. On the other hand, the fusion was detectable in every single blood draw.

"This patient had optimal surgery," Martignetti said, with no discernable traces of the tumor left behind, yet it was still there and detectable, and she ultimately recurred and succumbed to her disease.

Martignetti said that the team is now refining its sequencing protocol. Since the majority of patients have TP53 mutations, it will continue to do full-gene sequencing of that one, he said.

However, the group is also looking to develop a customized AmpliSeq panel that is more specific to gynecological cancers. The goal is to design a panel such that each patient will have at least four mutations that can then be tracked in subsequent noninvasive blood-based analyses via ddPCR using the RainDance system and Taqman reagents.

To design such a panel, the group is doing exome and transcriptome sequencing to figure out which mutations to include. Martignetti said that while such a comprehensive approach is great for discovery purposes, it would be too costly and time consuming to do clinically for every patient, with 200x exome sequencing running around $2,000.

Although it is still early in the trial, Martignetti said that it appears that ctDNA may be a better biomarker than CA125. For instance, he said, when ctDNA levels fall to zero post surgery and chemotherapy, patients have a better outcome. In addition, it seems that increases in ctDNA levels are detectable prior to other clinical measures of occurrence. Even in patients for whom CA125 levels are predictive of disease, ctDNA levels still appear to be an earlier predictor of recurrence, oftentimes months before CA125 levels increase.

In one case, ctDNA was a better predictor of drug response, Martignetti said. He described one woman whose CA125 levels were not changing and for whom the drug side effects were severe. However, ctDNA levels were falling dramatically and there were other clinical signs that she was responding to the medication.

Martignetti said that the woman, who was about to stop the medication because of the side effects and because her CA125 level was not declining, decided to stay on the medication. Her tumor shrank. However, she ultimately succumbed to the disease because she stopped taking the drug due to the severe adverse effects.