The 2026 European Society for Medical Oncology (ESMO) Congress was held in Madrid, Spain, from October 23 to 27. At this year's congress, Professor Jihong Liu's team from Sun Yat-sen University Cancer Center presented two studies in cervical cancer.

The Phase III DUBHE-C-301 study (1240O) reported, for the first time, results of first-line treatment with the PD-1/CTLA-4 bispecific antibody QL1706 in combination with chemotherapy, with or without bevacizumab, for persistent, recurrent, or metastatic cervical cancer. Both co-primary endpoints, progression-free survival (PFS) and overall survival (OS), were positive. Meanwhile, the Phase II study of SHR-A2102 (1244RO) demonstrated the efficacy and safety of a novel Nectin-4-targeted antibody-drug conjugate (ADC) in recurrent or metastatic cervical cancer. In addition, the preliminary findings of the STEP study, a cross-sectional survey of clinical practice in HER2 testing and treatment for endometrial cancer in China led by Professor Liu, were also discussed during the interview.

At the congress, Oncology Frontier interviewed Professor Jihong Liu about the evolving first-line treatment landscape with bispecific antibodies, the role of ADCs in cervical cancer, the current status of HER2 testing, and key insights from this year’s meeting. The following is the full interview.


Oncology Frontier: What were the Phase III results of QL1706 plus chemotherapy, with or without bevacizumab, as first-line treatment for persistent, recurrent, or metastatic cervical cancer? What improvements did this bispecific antibody combination offer over current standards of care?

Professor Jihong Liu: This study, DUBHE-C-301, was presented by me as an oral presentation at the ESMO Congress. First, regarding the primary endpoints, the study met both co-primary endpoints of PFS and OS. Patients in the QL1706 group had significantly better PFS and OS than those in the placebo group, meeting our expectations.

Regarding the secondary endpoints, the objective response rate (ORR) was substantially higher in the QL1706 group than in the placebo group. Particularly noteworthy was the considerably longer median duration of response (DoR): 16.1 months in the treatment group versus only 6.8 months in the control group, approximately 2.4 times longer. This is consistent with the characteristics of immunotherapy: it can not only improve response rates, particularly complete response rates, but also make responses more durable.

Furthermore, there was no significant difference between the two groups in subsequent anticancer treatments received after disease progression. This suggests that the primary benefits observed in this study were indeed attributable to the addition of the bispecific immunotherapy combination itself, rather than being confounded by subsequent treatments.

According to the study data, as of January 16, 2026, a total of 492 patients had been randomized: 243 to the QL1706 group and 249 to the placebo group. Of these patients, 80.9% had a PD-L1 combined positive score (CPS) of ≥1, and 48.4% had previously received concurrent chemoradiotherapy. Bevacizumab was administered to 77.4% and 75.1% of patients in the two groups, respectively.

According to blinded independent central review (BICR), median PFS was 12.35 months in the QL1706 group and 7.75 months in the placebo group. Median OS was 30.95 months and 21.78 months, respectively. The PFS and OS benefits were consistent across subgroups defined by age, distant metastasis, prior concurrent chemoradiotherapy, PD-L1 expression, and bevacizumab use.

Regarding safety, the overall incidence of major toxicities was broadly comparable between the two groups. The incidence of treatment-related adverse events (TRAEs) was 98.8% versus 99.2%, while the incidence of Grade ≥3 TRAEs was 80.6% versus 75.9%. Immune-related adverse events were significantly more frequent in the immunotherapy group, but were within the expected range, and no new adverse events were observed.

This study offers several implications for current standard treatment. The design of DUBHE-C-301 was almost identical to that of KEYNOTE-826, which established the basis for current standard-of-care recommendations. Chemotherapy, with or without bevacizumab, combined with pembrolizumab is currently a first-line standard for advanced recurrent cervical cancer.

Following KEYNOTE-826, DUBHE-C-301 is another prospective randomized Phase III study demonstrating that adding immunotherapy to chemotherapy, with or without bevacizumab, can improve first-line survival outcomes. Moreover, compared with KEYNOTE-826, the PFS benefit in this study was more pronounced. This may suggest that bispecific antibodies could potentially achieve better efficacy than monoclonal antibodies in the first-line setting.

Second, this study suggests that survival benefits can also be achieved in patients with low or even negative PD-L1 expression (CPS <1), as well as in patients who do not receive bevacizumab. The survival benefits in these populations were also encouraging.

Third, at this meeting, the principal investigator of the BEATcc study presented its final survival results. I noticed that the median OS in the placebo group of our study was 21.78 months, almost identical to the 22.7 months reported for the placebo group in the final BEATcc analysis. Although these were not head-to-head studies, the nearly identical OS results in the control groups suggest that the control treatment was sufficiently comprehensive in both studies and that the quality of the studies was comparable. During the discussion following our presentation, Professor Nicoletta Colombo summarized this point by commenting that the study had a “clean design” and that its benefits were consistent with those observed in BEATcc.

A “clean design” means that the dual primary endpoints in our study were assessed by a blinded independent review committee, whereas BEATcc relied on investigator assessment. Evaluation by an independent review committee is more objective and further supports the credibility and reliability of our results.

Of course, we must also acknowledge that the OS follow-up in our study is still relatively limited. The median follow-up was just over 29 months, and the upper limit of the 95% confidence interval for median OS in the QL1706 group had not yet been reached. In addition, patients with CPS <1 (PD-L1-negative disease) represented a relatively small proportion of the overall population in the subgroup analysis. We must therefore interpret findings from these subgroups very cautiously. Subgroup analyses should not be directly extrapolated into definitive clinical conclusions.

The management of long-term survivors, as well as how this regimen should be combined or sequenced with ADCs, will require further follow-up and investigation.


Oncology Frontier: What were the efficacy and safety results of SHR-A2102 in recurrent or metastatic cervical cancer? How do you view the role of ADCs in the cervical cancer treatment landscape?

Professor Jihong Liu: SHR-A2102 is a novel Nectin-4-targeted ADC composed of a monoclonal antibody targeting Nectin-4, a cleavable linker, and a topoisomerase I inhibitor payload.

The results presented by our team came from a Phase II clinical study (NCT06654440) in patients with recurrent or metastatic cervical cancer. The study included two dose cohorts: 36 patients received 6 mg/kg and 14 patients received 8 mg/kg, for a total of 50 patients.

As of January 30, 2026, the median follow-up was 8.8 months. The confirmed ORR in the overall population was 40%, with rates of 47.2% in the 6 mg/kg cohort and 21.4% in the 8 mg/kg cohort. The disease control rate (DCR) reached 88%. Median PFS was 5.9 months, reaching 7.0 months in the 6 mg/kg cohort, while the median DoR was 7.1 months.

The main toxicities were hematologic. The incidence of Grade ≥3 TRAEs was 58.0%, with neutropenia, anemia, and leukopenia being the most common events. No treatment discontinuations due to TRAEs were observed. Considering both efficacy and safety, the 6 mg/kg cohort may be selected as the recommended dose for further studies.

It is worth noting that 78% of patients in this study had previously received immune checkpoint inhibitors, and 66% had received bevacizumab. A substantial proportion had already undergone second-line or later treatment. In this population, where treatment options are very limited, achieving a 40% response rate with monotherapy is a very encouraging result.

A confirmatory Phase III study (NCT07418749) is currently underway. In addition, the cervical cancer indication for SHR-A2102 was included in the Center for Drug Evaluation of China’s National Medical Products Administration’s Breakthrough Therapy designation program in April 2026.

Regarding the prospects for ADCs in cervical cancer, I personally believe that this field is still at a very early stage and requires continued exploration. Based on the evidence currently available, I do not support moving ADCs directly into first-line or even initial treatment. ADCs should initially be positioned in later lines of therapy.

We need more evidence, particularly from large-scale Phase III randomized controlled trials. We must conduct rigorous studies to identify ADC regimens that genuinely improve survival while maintaining good tolerability and preserving patients’ quality of life.

As for Nectin-4 as a target, the toxicity of ADCs directed against it appears relatively low compared with that of some ADCs we have used previously. Of course, the number of patients treated so far remains very limited—only 50 in this study—and the observations regarding subjective toxicity have been relatively favorable. At the current stage, however, I have not yet seen evidence that ADCs can surpass conventional chemotherapy. There is still some distance to go before they can be applied in first-line treatment. This is my personal view.


Oncology Frontier: Your team led a cross-sectional survey of clinical practice in HER2 testing and treatment for endometrial cancer in China. What issues did the study reveal regarding access to HER2 testing?

Professor Jihong Liu: The STEP study was a questionnaire-based survey focusing primarily on HER2 testing. We now have many drugs targeting HER2, and testing is a prerequisite for using these agents. Therefore, we wanted to understand the current landscape of HER2 testing in China: How familiar are physicians with HER2? What are their attitudes and perceptions? Although physicians consider HER2 important, do they believe that testing methods and costs are acceptable to patients? Our goal was to assess the current status and accessibility of HER2 testing in gynecologic oncology in China.

The questionnaire survey revealed that a considerable proportion of physicians still do not attach sufficient importance to HER2. For patients with gynecologic cancers, many physicians do not recommend HER2 testing at initial diagnosis or during initial treatment, but instead wait until the disease recurs before recommending testing. Even when patients have already undergone HER2 testing during initial treatment, 90% of physicians would still recommend repeating the test after recurrence. This is the current situation, although I find it difficult to speculate on the reasons behind physicians’ decisions.

Another issue is that HER2 testing places very high demands on specimen quality. Freshness is particularly important. If a specimen has been stored for more than one or two years before testing, the results are often inaccurate. Tumor specimens are generally paraffin-embedded, and we often ask patients to submit tissue sections for testing. However, once prepared, these sections can only be stored for a relatively short period, generally no longer than one month. Testing after more than a month may also produce inaccurate results.

At the same time, there is still no consensus on the pathological scoring criteria for HER2 in gynecologic cancers. In clinical practice, the two existing scoring systems are those used for gastric cancer and breast cancer. In gynecologic oncology, we currently refer to the gastric cancer scoring criteria, but whether this is appropriate remains to be determined through further evidence.

Based on the survey findings, we plan to conduct more targeted studies. For example, we may focus on a specific gynecologic cancer and compare one or two testing methods to determine which scoring criteria are most appropriate for that disease.

Overall, many issues surrounding HER2 testing in gynecologic oncology remain unresolved and deserve continued attention and efforts to address them.


Oncology Frontier: Which study at this year’s ESMO Congress left the deepest impression on you, and why?

Professor Jihong Liu: The study that impressed me most at this year’s ESMO Congress may not actually have been in gynecologic oncology. The ESMO Congress covers all tumor types rather than focusing exclusively on gynecologic cancers. For gynecologic oncologists like us, it provides an excellent opportunity to learn from other disciplines.

While reviewing the conference program, I came across a study that particularly interested me: the SANO study, which investigated neoadjuvant treatment for esophageal cancer and presented its long-term follow-up results at this meeting.

The reason it made such a strong impression on me is that I am currently preparing to launch a study in patients with advanced ovarian cancer, comparing surgery versus no surgery following neoadjuvant chemotherapy. The design and findings of SANO have given me considerable inspiration.

Different medical disciplines can learn from and advance one another. By attending major congresses such as ESMO, we can draw on the experience and research foundations of other specialties. This is extremely valuable for advancing both research and clinical practice in gynecologic oncology.