番茄社区

SAN ANTONIO -- Negative results with a blood-based test were perfectly accurate in ruling out cancer recurrence in patients with HPV-related oropharyngeal squamous cell carcinoma (OPSCC) in the post-treatment surveillance setting, a prospective study found, although positive results were not so accurate.

At a mean follow-up of 19.8 months, no levels of circulating tumor HPV DNA were detected at any point in 70 of 89 OPSCC patients who had been treated with definitive chemoradiation, and all 70 remained disease free, yielding a negative predictive value of 100%, reported Bhisham Chera, MD, of UNC Lineberger Comprehensive Cancer Center in Chapel Hill, at the American Society for Radiation Oncology (ASTRO) meeting.

The remaining 19 patients tested positive for circulating tumor HPV DNA (median 75 copies/mL) at a median of 16.7 months from the time of treatment, with eight going on to have biopsy-proven recurrence, for a positive predictive value (PPV) of 42%, according to the findings presented during a late-breaking ASTRO session.

"I think the key message is it can make our ability to surveil patients more effective and also possibility detect cancer sooner than we normally do now, which may allow for the salvage to be more effective," Chera told 番茄社区.

Co-investigator Gaorav Gupta, MD, PhD, also of UNC Lineberger, told 番茄社区 that the costs of surveillance scans can be prohibitive.

"We can't do them every visit, so if you had a blood test that would trigger a scan, then maybe we could detect recurrence at an earlier stage where perhaps surgery or radiation and/or systemic therapy might have a better effect," he said.

Of the eight patients with confirmed recurrence, one patient had recurrence of local disease, four had distant disease, and three had both. Among the 11 patients with a positive HPV DNA test, the positive signal was cleared in four patients while the other seven received close monitoring.

Chera noted that patients who test positive are not necessarily treated. "If there's no clinical or radiographic exam findings we would not treat those patients," he said. "We continue to check them, and hopefully their immune system will clear that out like they did in those four patients."

"To see something on a scan, you need millions and millions of cells," Richard Bakst, MD, of Icahn School of Medicine at Mount Sinai in New York City, told 番茄社区. "Often patients get PET scans, but to see something on a PET scan, which is incredibly sensitive, you still need a large volume of cancer."

Bakst, who was not involved in the research, noted that liquid biopsies will no doubt have a place in clinical care but that there's still work to be done with tests such as this one. "I think the lower PPV at this point suggests that it's not quite ready for prime time."

"If you're detecting a result or a signature, what do you do with that patient?" he said. "I think that's the patient we really need to think about."

With the caveat that it may not be the perfect analogy, Bakst likened the current utility of liquid biopsies to PSA testing for surveillance in prostate cancer patients.

"It's sort of this gray area of medicine," he said. "You see this PSA rising but you don't know what to do with that -- you know there's probably cancer somewhere, you don't know where."

Bakst said that one of his philosophies as a clinician is: "Don't order a test if you don't know what to do with it," as otherwise it can simply create anxiety for patients.

All 89 patients in the study had biopsy-proven HPV-related disease. While all received definitive chemotherapy, 78 received a de-intensified therapy (60 Gy) and 11 received standard therapy (70 Gy).

Patients had follow-up care every 2 to 4 months, and then every 6 months for years 3 to 5. Chest imaging (CT, PET/CT, and x-ray) was conducted every 6 months and additional imaging was obtained in patients with a positive HPV DNA test. The multi-analyte digital PCR assay used was able to detect high-risk HPV strains (16, 18, 31, 33, and 35). In all, roughly 1,000 blood tests were conducted among the 89 patients.

Chera said that a larger trial will be needed to "show that the negative predictive value is really 100% or very close to that" and that other possible research directions could involve the inclusion of patients with other HPV-related disease, such as cervical or anal cancers, to increase the usefulness of the test.

Comment