乌鸦传媒

HPV-positive circulating tumor (ct)DNA after chemoradiation (CRT) for cervical cancer had an independent association with worse outcomes, suggesting potential as a biomarker to guide decision-making, a prospective, multicenter study showed.

Patients with detectable HPV ctDNA at the end of CRT had significantly worse 2-year progression-free survival as compared with undetectable HPV ctDNA (77% vs 51%, P=0.03). The strength of the association increased over time and remained statistically significant for HPV ctDNA detected 4-6 weeks and 3 months after CRT.

By multivariable analyses, HPV ctDNA after CRT by digital polymerase chain reaction (dPCR) or next-generation sequencing (HPV-seq) remained independently associated with worse 2-year PFS, reported Kathy Han, MD, of Princess Margaret Cancer Centre in Toronto, and co-authors in the (JCO).

"HPV-seq enables sensitive HPV detection and genotyping directly from plasma," the authors concluded. "HPV ctDNA testing can be used to identify, as early as at the end of CRT, patients at high risk of recurrence in future treatment-intensification trials."

The findings have implications for development of more effective therapies for high-risk cervical cancer, JCO associate editor Gini F. Fleming, MD, of the University of Chicago, noted in a summary of the article's relevance.

"Useful salvage therapy is not currently available for the majority of relapsed cervical cancer patients," wrote Fleming. "These results should be useful in selecting the highest risk patients for the development of treatments to prevent clinical recurrence."

About 30-40% of patients with locally advanced cervical cancer relapse after CRT. Clinical prognostic factors, such as stage and lymph node status, do a poor job of predicting relapse, the authors noted in their introduction.

Most cervical cancers are caused by HPV infection, which provides a convenient genetic marker of cancer-derived DNA that could be used to identify residual disease burden in plasma. In a previously reported , Han and colleagues found that detectable HPV ctDNA at the end of CRT was associated with worse PFS. The study also suggested that HPV-seq might outperform dPCR.

The investigators hypothesized that HPV ctDNA might identify patients with cervical cancer at high risk for recurrence after CRT. They designed a prospective study to validate HPV ctDNA as a means of early detection of molecular residual disease (MRD).

The study involved patients from four Canadian centers treated for stage IB2-IVA squamous cell cervical cancer from 2017 to 2022. Investigators compared HPV ctDNA and PET imaging to evaluate patients for residual disease following CRT. Patients were examined every 3-4 months during the first 2 years of follow-up and then every 6 months during years 3-5.

PET imaging was performed 3-4 months after CRT. Investigators analyzed plasma samples by HPV-seq and dPCR immediately after CRT, at 4-6 weeks, and 3 months after CRT. Patients enrolled after April 2019 were also evaluated by FDG-PET/CT imaging. The primary outcome was PFS at 2 years.

Data analysis included 70 patients, a sample size determined by a 2-year PFS of 70% from a large institutional series and the assumption that 30% of patients would have detectable HPV ctDNA following CRT. The patients had a mean age of 54. The most common FIGO disease stages were IIB (30%), IIIC1 (24%), IIA and IVA (11%), and IIIC2 (10%).

The lowest level of detectable HPV ctDNA was 0.18 copy/mL by dPCR and 0.023 copy/mL by HPV-seq. The median lead time from detection of HPV ctDNA to identification of relapse by examination/radiologic imaging was 5.9 months for both dPCR and HPV-seq.

Patients with detectable HPV ctDNA by dPCR had significantly worse 2-year PFS, regardless of when HPV ctDNA was detected:

• Immediately after CRT: 77% vs 51%, P=0.03
• 4-6 weeks after CRT: 82% vs 15%, P<0.001
• 3 months after CRT: 82% vs 24%, P<0.001

Results were similar with HPV-seq:

• Immediately after CRT: 87% vs 53%, P=0.009
• 4-6 weeks after CRT: 79% vs 39%, P<0.001
• 3 months after CRT: 85% vs 26%, P<0.001

The accuracy of MRD detection by HPV ctDNA did not differ significantly between HPV-seq and dPCR for predicting PFS at any of the measurements. In the subgroup of 44 patients who underwent FDG-PET/CT imaging at 3 months, prediction of 2-year PFS also did not differ significantly from the two HPV ctDNA detection methods.

"To our knowledge, this is the first study that prospectively validates HPV ctDNA as a tool for early detection of residual disease after CRT in patients with locally advanced cervical cancer," the authors wrote in their discussion of the findings. "Persistent HPV ctDNA after CRT was independently associated with inferior PFS in this prospective validation study at all time points, thus confirming our previous results."

"There was no statistically significant difference between the performance of HPV-seq and dPCR in MRD detection nor between the timepoints that were evaluated, highlighting the benefit of early identification of patients with MRD at risk of relapse."

The finding that the two methods of detecting HPV ctDNA had similar accuracy was unexpected, the authors continued, despite the greater sensitivity of HPV-seq. They speculated that "modest numerical improvements in sensitivity" favoring HPV-seq were offset by reduced specificity of HPV-seq versus dPCR.

Comment