HPV ctDNA Leads the Field in Cervical Cancer MRD Monitoring
Liquid biopsy is moving closer to the bedside in gynecologic oncology, but the evidence is not evenly distributed across tumor types. In cervical cancer, post-treatment HPV circulating tumor DNA has the clearest disease-specific prospective support for measuring minimal residual disease and anticipating recurrence. Persistent molecular detection after therapy tracks closely with later relapse, giving the assay a biological logic that fits the known HPV-driven origin of the disease: residual infected malignant clones can continue shedding viral tumor DNA even when imaging and pelvic examination remain unrevealing.
A single negative test, however, cannot be treated as a clean bill of health. Sensitivity remains only moderate, and false-negative results leave room for residual disease that falls below assay thresholds, is spatially confined, or sheds poorly into blood. The review draws a firm clinical boundary here: negativity alone is not enough to justify treatment de-escalation.
That caution matters at the bedside. A patient with apparent remission may still harbor microscopic disease after chemoradiation or surgery, and a biomarker that misses a meaningful fraction of those cases cannot safely guide less intensive follow-up or reduced therapy. The strongest use case today is therefore risk refinement, not treatment withdrawal.
Endometrial ctDNA Has Prognostic Signal but Uneven Validation
Endometrial cancer presents a different technical and biological problem. Postoperative ctDNA has shown prognostic value across an 11-study, 1298-patient meta-analysis and in additional cohorts, suggesting that molecular relapse detection may eventually help identify patients at higher risk of recurrence before standard surveillance turns abnormal. The signal is encouraging, but the field remains fragmented by assay design, timing of sampling, and differences in how residual disease is defined.
Postoperative positivity appears to carry stronger prognostic weight than preoperative detection. This makes clinical sense: after the tumor has been removed, any detectable tumor-derived DNA is more likely to represent residual malignant burden rather than baseline disease volume alone. Preoperative ctDNA, by contrast, reflects tumor load before definitive therapy and is less specific for post-treatment risk stratification.
The problem is not simply one of numbers. Assay heterogeneity limits direct comparison across studies. Tumor-informed strategies, fixed disease-specific panels, digital PCR, and error-corrected sequencing all measure different biological and analytical slices of the same problem, and each brings distinct tradeoffs in sensitivity, turnaround time, breadth, and cost. Without a unified technical standard, the field can show prognostic associations without yet proving which test should be used, when it should be drawn, or how results should change management.
Local Recurrence Detection Is More Mature in Cervicovaginal and Urine Methylation Studies
Cervicovaginal and urine-based DNA methylation assays add another layer to the MRD discussion. These approaches have produced preliminary evidence for detecting established recurrence, with the strongest signals appearing in local relapse. The compartment matters: when a tumor recurs near the cervix, vagina, or lower urinary tract, local shedding into adjacent fluids may make molecular detection easier than plasma-based surveillance.
Still, these studies remain early. They have not yet delivered the sort of prospective molecular lead time that would prove recurrence can be detected earlier in a clinically actionable way, and they have not established that earlier molecular detection improves outcomes rather than simply identifying recurrence sooner. A test that detects disease after it is already visible to routine methods may still be useful, but it does not yet fulfill the promise of intervention before clinical relapse.
For now, methylation assays look more like an enriched detection strategy for known recurrence than a validated MRD tool for post-treatment surveillance.
Why the Technology Matters as Much as the Biomarker
The liquid biopsy field in gynecologic cancer is being shaped by platform choice as much as by tumor biology. Digital PCR offers high analytic sensitivity for a small number of known targets, making it suitable when a specific mutation or viral sequence is already established. Tumor-informed assays can improve specificity by tailoring the test to the patient’s tumor profile, which is attractive for MRD monitoring after definitive treatment. Error-corrected sequencing extends breadth and can identify rare variants with greater confidence, but the added complexity can make implementation harder in routine practice.
Broad pan-cancer plasma profiling sits in a different category. It has value in advanced disease and in settings where a comprehensive genomic picture is needed, but it is not yet the best fit for subtle post-treatment surveillance in endometrial or cervical cancer. MRD detection requires extreme analytical precision, because the target signal is faint and the consequences of both false reassurance and false alarm are substantial.
The biology also differs by specimen. Plasma may dilute low-volume disease, while cervicovaginal and urine samples can enrich for local shedding. That distinction helps explain why some assays perform better for local recurrence than for distant relapse, and why a negative blood test cannot exclude residual pelvic disease with confidence.
Exploratory Biomarkers Remain Hypothesis-Generating Only
Beyond ctDNA and methylation, the review addresses circulating tumor cells, extracellular vesicles, microRNAs, tumor-educated platelets, and fragmentomics. These platforms are intriguing, but they remain exploratory in this disease space. Each offers a different window into tumor-host interaction, and each faces practical barriers related to standardization, reproducibility, and biological interpretation.
A fragmentomics profile may reflect nucleosomal patterns from tumor cell death; microRNAs can mirror post-transcriptional regulation; platelet RNA signatures may capture systemic tumor signaling. None of these approaches has yet crossed the threshold from biologic plausibility to durable clinical utility in endometrial or cervical cancer MRD surveillance.
The gap is not conceptual. It is translational.
Prospective Trials Must Prove That Molecular Action Improves Outcomes
The central question is no longer whether liquid biopsy can detect tumor-derived material after treatment. In cervical cancer, HPV ctDNA already answers part of that question with meaningful prospective support. In endometrial cancer, perioperative ctDNA shows a strong enough prognostic signal to justify continued development. The real endpoint now is whether acting on a molecular result changes the course of disease in a way that benefits patients.
That means prospective trials with predefined intervention rules, not retrospective enthusiasm. Molecular positivity must be linked to a treatment pathway that can plausibly improve survival, reduce relapse, or refine surveillance intensity without excess harm. Until then, liquid biopsy belongs in the research setting or as an investigational adjunct to routine follow-up.
For patients and clinicians, the practical implication is restraint. Molecular relapse detection is advancing, but a positive test still needs a management framework, and a negative test cannot yet authorize less care. The field has reached the point where analytical success is no longer enough; clinical utility must come next.