Neoadjuvant rilvegostomig (AZD2936) plus fam-trastuzumab deruxtecan-nxki (Enhertu; T-DXd) produced high pathologic complete response (pCR) rates in patients with immune-positive response predictive subtypes (RPS) of high-risk HER2-negative breast cancer, according to results from the phase 2 I-SPY 2.2 trial (NCT01042379) presented at the
In patients with an immune-positive RPS, Block A treatment with rilvegostomig plus T-DXd alone, administered prior to any additional treatment blocks, yielded a pCR rate of 57% (95% CI, 35%-78%) in patients with hormone receptor (HR)–positive, HER2-negative, immune-positive disease (n = 12) and 52% (95% CI, 32%-71%) in patients with HR-negative, HER2-negative, immune-positive disease (n = 35). Across all treatment blocks, 34 of 47 immune-positive patients (72%) achieved a pCR; 62% of those pCRs occurred following Block A alone, enabling surgery without chemotherapy, and 97% were achieved by end of Block B without anthracycline/cyclophosphamide (AC).
I-SPY 2.2 is a multiple sequential treatment block, response-adaptive platform trial enrolling patients with MammaPrint high-risk, stage II/III breast cancer. The trial is designed to assign and personalize systemic therapy to maximize pCR rates without unnecessary toxicity.1,2
Patients in the experimental arm received rilvegostomig, an Fc-reduced anti–PD-1/TIGIT bispecific antibody, plus T-DXd in Block A, with response-adaptive escalation to Blocks B and C if needed prior to surgery.1 The concurrent control arm received standard taxane/AC-based neoadjuvant chemotherapy.
The primary endpoint was pCR rate after Block A alone and after the full treatment strategy vs the control regimens. Patients were stratified by RPS version 2.0 into four subtypes: HR-positive, HER2-negative, immune-positive (n = 12); HR-positive, HER2-negative, immune-negative (n = 47); HR-negative, HER2-negative, immune-positive (n = 35); and HR-negative, HER2-negative, immune-negative (n = 11).
A total of 105 patients were enrolled to receive rilvegostomig plus T-DXd arm had a median age of 48 years; 56% of patients had hormone receptor–positive disease; 45% had immune-positive disease; and 72% had a HER2 immunohistochemistry score of 1+/2+.
Rationale for the combination included preclinical data demonstrating that the anti–PD-1/TIGIT bispecific plus T-DXd improved tumor response compared with T-DXd plus a PD-L1 inhibitor. A prospective safety taskforce oversaw interstitial lung disease (ILD) monitoring using CT chest imaging every 6 weeks, pulmonary function tests, and 6-minute walk tests.
When evaluating rilvegostomig plus T-DXd, followed by standard chemotherapy blocks as needed, pCR rates were similar to concurrent controls across all RPS subtypes.
For immune-positive subtypes, estimated pCR rate was 66% (95% CI, 40%-87%) with rilvegostomig plus T-DXd vs 65% (95% CI, 35%-89%) with control in patients with HR-positive, HER2-negative, immune-positive disease (P = .51); and 72% (95% CI, 56%-85%) vs 70% (95% CI, 49%-86%) in patients with HR-negative, HER2-negative, immune-positive disease (P = .55). In the immune-negative subtypes, activity with block A treatment alone was low, with estimated pCR rates of 3% (95% CI, 0%-7%) for patients with HR-positive, HER2-negative, immune-negative disease and 8% (95% CI, 2%-21%) for those with HR-negative, HER2-negative, immune-negative disease; the full strategy remained comparable with controls.
A sensitivity analysis by HER2 IHC status suggested rilvegostomig plus T-DXd may improve pCR in patients with HER2 IHC 1+/2+, immune-positive disease vs control, with estimated pCR rates of 74% vs 62% (P = 0.82), a finding investigators noted warrants further evaluation.
Rilvegostomig plus T-DXd was tolerable in the intent-to-treat population. Interstitial lung disease (ILD) occurred in 14 patients (13.3%), with 12 events in Block A and 2 in Block B; ILD was primarily low-grade, with grade 3 or higher ILD occurring in 1.0% of patients. Per protocol, both rilvegostomig and T-DXd were discontinued upon any grade 1 or higher ILD, which led to treatment discontinuation for 8 patients in Block A. All patients with ILD recovered without recurrence.
Compared with controls, the regimen demonstrated decreased rates of neutropenia, anemia, musculoskeletal pain, peripheral neuropathy, and rash, while rates of fatigue, nausea, and constipation were comparable.
