A new cancer medicine separates median progression-free survival in two patient groups from 6.8 months to 16.4 months. That is news worth taking seriously. But we may still ask: will patients live longer? Will they feel better during treatment? What side effects must they endure to gain this period of disease control?
On October 23, 2026, the phase III HORIZON-Breast06 results for Hengrui Pharma's HER2 antibody–drug conjugate trastuzumab rezetecan were published in JAMA. The study was also presented at the ESMO congress in Madrid, providing a concrete starting point for these questions. [1]
A few days later, another announcement added a further dimension. BioNTech and DualityBio reported that a phase III study of another HER2 ADC had also met its primary endpoint. Yet they also judged that, based on the available efficacy and safety data, the potential patient benefit was limited in the current treatment landscape. [2]
The studies differ in their patients and designs, so they cannot be used to rank the two drugs. They can, however, help explain why a “positive result” deserves further reading.
First question: what specifically improved when the numbers improved?
Hengrui's medicine is called trastuzumab rezetecan in English, with the development code SHR-A1811. It is an antibody–drug conjugate, or ADC, linking an antibody that recognizes a particular target to a cytotoxic drug, with the aim of delivering the drug into the corresponding tumor cells.
The study addressed HER2-low advanced breast cancer. HER2 is a receptor protein on the cell surface. “Low” expression means there is less of this protein, rather than no recognizable target at all. ADC development has allowed researchers to explore treatment opportunities for this patient population anew. [1]
HORIZON-Breast06 enrolled 551 patients at 73 centers in China, comparing trastuzumab rezetecan with investigator's-choice chemotherapy. In the overall population, median progression-free survival was 16.4 months versus 6.8 months, with a hazard ratio for progression or death of 0.27. The paper also reported results separately for patients who had not received chemotherapy for advanced disease. This article uses the overall-population figures to avoid mixing results from different populations. [1]
Progression-free survival is generally measured from randomization to disease progression or death, whichever occurs first. It tells us how long the study regimen can control disease, but cannot by itself tell us how long patients ultimately live. A median is also a statistic describing a group; it cannot be assigned directly to each patient.
Delaying progression may mean that patients face a treatment switch, further tests, and new treatment arrangements later. That has value in itself. At the same time, longer disease control, better quality of life, and longer life each require their own evidence. Overall survival data in this study remain immature, so conclusions about extending life must wait. [1]
Second question: what treatment burden buys that time?
The safety section of the same paper provides another set of numbers that must be read alongside efficacy: grade ≥3 treatment-related adverse events occurred in 84.1% versus 67.9%, and treatment-related serious adverse events in 37.7% versus 13.7%. [1]
These percentages use different definitions. “Grade ≥3” primarily reflects the intensity of an adverse event; “serious adverse event” is determined by consequences such as hospitalization or a threat to life. An event can fall into both categories. The rates cannot be added or used interchangeably.
For patients, knowing the percentages is only a first step; what actually happened matters too. Neutropenia, anemia, and lung inflammation affect the body and daily life differently. Whether they can be managed with monitoring, supportive care, treatment pauses, or dose reductions also changes the treatment experience.
Event rates are also related to treatment duration. Patients treated for longer have more time in which adverse reactions can occur. One cumulative percentage therefore cannot capture all of safety. Equally, a longer treatment duration does not make the burden that actually occurred irrelevant.
This turns “several more months of control” into a fuller question: during those months, how many symptoms and tests did patients experience? Did they need hospitalization, dose reductions, or treatment discontinuation? How far could they maintain their usual lives?
Third question: why were the companies cautious after the other drug met its endpoint?
On October 26, BioNTech and DualityBio announced topline results from the global phase III DYNASTY-Breast02 trial of DB-1303, also known as BNT323 or trastuzumab pamirtecan. They reported a statistically significant but modest improvement in PFS versus chemotherapy, with overall survival data still immature. [2]
The announcement also noted a higher incidence of interstitial lung disease than in earlier trials of the drug. Most cases were grade 1 or 2, but a small proportion of patients experienced serious and fatal events. Considering the available efficacy and safety together, the companies judged the potential patient benefit to be limited in the current treatment landscape for hormone receptor-positive, HER2-low metastatic breast cancer, and said they would discuss next steps. [2]
The announcement did not provide the complete PFS effect size or the specific incidence of lung disease. We therefore cannot yet calculate the benefit–risk balance precisely, or infer that the entire program has been discontinued. Its real lesson is that the existence of a statistical difference and whether that difference has sufficient clinical value need separate assessments.
Can design differences explain different results between HER2 ADCs?
Both medicines target HER2, carry topoisomerase I inhibitors, and use enzyme-cleavable linkers containing the GGFG tetrapeptide. Hengrui's payload, however, is SHR169265, while DualityBio's is P1003. Their average numbers of payload molecules per antibody—the drug-to-antibody ratio, or DAR—are approximately 6 and 8, respectively. Hengrui's preclinical work compared different DARs and selected 6 to balance antitumor activity and toxicity. [3][4]
These differences offer clues for further research. Payload potency, cell permeability, release and clearance, and the dose and exposure of the whole molecule may all affect performance. But fewer payload molecules do not, by themselves, establish greater safety. The same target and enzyme-cleavable sequence likewise do not guarantee that the entire molecules have the same properties.
Design differences can help generate hypotheses about why clinical performance differs; establishing the specific cause requires direct comparisons and the appropriate evidence. The difference between these trials therefore cannot simply be attributed to “6 payloads being better than 8,” or used to declare a platform the winner.
The same target does not make two sets of results directly rankable
Hengrui's study included both hormone receptor-positive and hormone receptor-negative patients, and allowed patients with no prior chemotherapy for advanced disease or one prior line. DualityBio's study enrolled hormone receptor-positive patients who had not received chemotherapy for advanced disease. The trials were conducted in different regions, and the specific chemotherapy options in the control arms were not identical. [1][2]
Patients' previous treatments and the extent of their disease may affect subsequent efficacy. Even when two trials use the same endpoint, each observed result depends on these conditions.
Hengrui's result therefore supports efficacy in the setting studied, but does not directly establish superiority over DualityBio's medicine. The limited benefit in DualityBio's study likewise cannot be generalized to every HER2 ADC, all patients, or other indications.
Final question: what other choices do patients have today?
After a study demonstrates that a drug is better than chemotherapy, physicians still need to decide where it belongs among existing treatments. Both trials used chemotherapy as the comparator, rather than a direct comparison with another HER2 ADC. [1][2]
Actual treatment choices also depend on whether patients can access other effective medicines, which ADC classes they have already received, which regimen suits their physical condition, and what options remain after toxicity or disease progression. The treatment landscape changes with new evidence, and new medicines need to find a place within it.
“Meeting a phase III endpoint” provides one important answer; a patient's treatment choice requires several answers together. Hengrui's study provides important evidence of disease control while raising questions about safety and long-term survival. DualityBio's announcement reminds us that even when a statistical endpoint is met, overall benefit still needs assessment alongside treatment burden and existing options.
The next time we encounter exciting news about a new medicine, we can linger a little longer: which measure improved, by how much, at what cost, and which questions are still waiting for answers? Asking these questions gradually clarifies the distance between good news and real life.
Main sources and further reading
Fan Y, Liu Y, Wang Y, et al. October 23, 2026: Full report of the HORIZON-Breast06 randomized clinical trial. Published in JAMA, examining trastuzumab rezetecan in HER2-low advanced breast cancer. The original includes eligibility criteria, efficacy in different populations, safety, and study limitations. DOI: 10.1001/jama.2026.21372.
BioNTech/DualityBio, October 26, 2026: Original announcement of global phase III DYNASTY-Breast02 topline results. Consult the companies' own wording on the magnitude of efficacy, pulmonary safety, and next steps. This article does not fill in specific values that the announcement had not disclosed.
Zhang T, et al. Original research on Hengrui's SHR-A1811 molecular design and DAR optimization. PLOS One, June 26, 2025. DOI: 10.1371/journal.pone.0326691. Provides information on the payload, linker, and preclinical optimization.
Early clinical study abstract for DualityBio's DB-1303: molecular composition and DAR. Cancer Research, 2023, abstract CT247. Cited here for molecular composition; early clinical results are not substituted for evidence from the current phase III trial.
Information current to October 27, 2026. This article is intended to aid understanding of research evidence and does not constitute a direct comparison or ranking of the two medicines. Specific treatment choices should be discussed with an oncologist in light of the patient's condition.
This is Before the Conclusion.
What does a news story, a number, or a seemingly excellent result actually mean? Follow Before the Conclusion as we start with medicine and science, examine the reasons and tradeoffs behind the evidence, and keep asking questions.
Text and image use
The text of this article was originally written by Before the Conclusion. You are welcome to share its link. Without permission, please do not copy, republish, repost screenshots, or use it commercially. Any accompanying illustrations provided separately are subject to the same restrictions and may not be used on their own without permission. Image production and source information are given in the respective captions.