A drug pairing let many breast-cancer patients skip the harshest chemo

A new pairing of two targeted drugs, given before surgery, cleared detectable breast cancer in just over half of a selected group of patients — and, more unusually, let most of them skip the harshest part of standard chemotherapy [s1]. The result comes from one arm of I-SPY2.2, a long-running trial built to test promising drugs quickly and match them to the patients most likely to benefit [s1][s2].

High-risk, early-stage breast cancer is usually treated with months of multi-drug chemotherapy, often including anthracyclines such as doxorubicin — effective drugs, but ones that carry heart and long-term safety costs. The appeal of a more targeted approach is not only better response but less exposure to that toxicity. I-SPY2.2 is designed to find exactly those trade-offs, enrolling patients at 30 US sites and assigning them adaptively as data accumulate [s1][s2].

What the trial did

This arm tested trastuzumab-deruxtecan, an antibody-drug conjugate that targets even low levels of the ERBB2 (HER2) protein and delivers a chemotherapy payload directly to cancer cells, combined with rilvegostomig, a bispecific antibody aimed at two immune checkpoints, PD-1 and TIGIT [s1]. The combination, abbreviated RT-DXd, was given as the first of up to three sequential treatment blocks: block A was RT-DXd, block B a taxane-based regimen with or without immunotherapy, and block C doxorubicin-cyclophosphamide with or without immunotherapy [s1]. Patients moved through the blocks according to how their tumour responded and which molecular subtype it belonged to [s1].

Enrolment was limited to high-risk stage II or III disease that was ERBB2-negative, including tumours with low or ultralow ERBB2 levels, and patients were grouped by hormone-receptor status and by response-predictive subtypes [s1]. The main measure was pathologic complete response — no residual invasive cancer found in the breast or lymph nodes at surgery [s1].

What it found

In all, 182 people took part: 88 were assigned to a contemporary control arm, and 108 of the 289 remaining participants (37%) to RT-DXd, leaving 105 and 77 evaluable patients after some did not receive their assigned treatment [s1]. The mean age was 48.1 years in the RT-DXd group and 50.9 in the control group [s1].

Among the 105 patients treated with RT-DXd in block A, the modelled complete-response rate reached 54% in tumours classed as immune-positive — enough to meet the trial's predefined threshold for success [s1]. The signal held across hormone-receptor status: 57% in hormone-receptor-positive, immune-positive disease and 52% in hormone-receptor-negative, immune-positive disease [s1].

The more striking finding is what those responders avoided. Across all treatment blocks, overall complete-response rates were similar to the control regimens — RT-DXd did not beat standard therapy on the headline number [s1]. But among immune-positive patients who achieved a complete response, 21 (62%) needed only block A, and 33 (97%) never received anthracycline-based chemotherapy at all [s1]. For a subset of patients, in other words, two targeted drugs did the work that normally takes a full chemotherapy course.

The safety picture required active management. Enrolment was paused twice to review interstitial lung disease — inflammation of the lung tissue, a known risk of trastuzumab-deruxtecan — and resumed after the safety monitoring board signed off [s1]. The most common side effects were fatigue and nausea. Serious events included pulmonary embolism and interstitial lung disease, each in 3 of 105 patients (2.9%), and lung inflammation of any grade occurred in 13 (12.4%), including one severe case, picked up early through repeated chest scans [s1].

How to read it

The honest reading is narrower than the headline rate suggests. This was a phase 2 result in a biomarker-selected slice of patients, measuring tumour clearance at surgery rather than survival, and the overall response rate was no better than standard care [s1]. What makes it notable is the de-escalation: an immune signature identified patients for whom a targeted pair could replace anthracyclines, which matters for people who would otherwise carry that toxicity for decades [s1].

Pathologic complete response is a useful early signal but not a guarantee of longer life, and the lung-inflammation rate is a reminder that "targeted" is not the same as "gentle." The trial's own design — adaptive, biomarker-driven, and continually analysed — also means these are modelled estimates meant to guide the next, larger study rather than settle practice [s1][s2].

What to watch

The investigators frame RT-DXd as worth testing further in biomarker-selected patients, and that is the open question: whether sparing immune-positive patients from anthracyclines preserves long-term outcomes as well as it preserves the complete-response rate. Confirmatory trials with survival endpoints, and a clearer definition of who counts as "immune-positive," will decide whether this becomes a real path to less chemotherapy.

This article describes research and is not medical advice. Breast-cancer treatment decisions belong with a specialist oncology team.