To Roche, a phase 3 flop in first-line breast cancer does not erase the megablockbuster potential of its oral SERD drug giredestrant even as competition mounts.
In fact, results from the failed persevERA trial in first-line HR-positive, HER2-negative metastatic breast cancer have given Roche confidence in a new study that will test the combination of giredestrant with a CDK4/6 inhibitor for the adjuvant treatment of early-stage breast cancer, Roche’s deputy chief medical officer, Stefan Frings, M.D., Ph.D., told Fierce in an interview.
“What we still see is a quite attractive separation of the curve […] You will clearly see there’s a treatment effect,” Frings said of progression-free survival (PFS) data from persevERA. “This convinces us, because in the adjuvant setting, we treat for five years, and if we move a few percentage points upwards, you would have already a positive treatment effect in the adjuvant setting.”
Frings wouldn’t disclose any details of the upcoming new adjuvant study, saying it’s still in the design phase.
Before the persevERA result, Roche already reported a positive readout from the lidERA trial of giredestrant monotherapy in the adjuvant setting.
The surprise first-line flop
After the lidERA hit in November, the persevERA miss came as a big surprise.
According to results presented at the American Society of Clinical Oncology 2026 annual meeting, the combination of giredestrant and Pfizer’s CDK4/6 inhibitor Ibrance (palbociclib) showed a numerical 11% PFS improvement by investigator analysis compared with Novartis’ aromatase inhibitor Femara (letrozole) and Ibrance, failing to meet statistical significance.
The giredestrant regimen’s 33.1-month median PFS was 4.9 months longer than that of the control arm. Because metastatic patients are treated until disease progression, half of them will receive giredestrant for less than three years, or just about half of the five-year duration intended for early-stage therapy.
The PFS curve separation that Frings discussed started to emerge around 16 months. At two years, 59.7% of patients in the giredestrant arm and 57.3% in the control group remained progression-free. By three years, the PFS rates were 45.8% and 41.9%, respectively.
persevERA primary endpoint: Investigator-assessed PFS (Nicholas Turner, et al./ASCO 2026)In contrast, giredestrant struggled to show any overall survival advantage, as the hazard ratio logged at 1.03, meaning death risk was marginally higher for the Roche drug, although the data were immature with a median follow-up of about 52 months. Three-year OS rates were the same at 74.1% for the two arms, and median OS was not reached in either arm.
“The reason why this may not have worked is that certainly CDK4/6 [inhibitors] are very potent agents,” Frings said, “and that you don’t see early on what is really an effect of an endocrine manipulation. A dead cell is dead, and you can’t kill it better when it’s already dead.”
On the safety side, adverse events (AEs) were generally comparable between the two treatment groups but were more frequent for giredestrant. Serious AEs were recorded in 23.9% of giredestrant patients, versus 18.8% in the control arm. Treatment-related adverse events led to one (0.2%) death and two (0.4%) deaths between the groups, respectively.
Roche has a second first-line study, pionERA, which is testing giredestrant alongside CDK4/6 inhibitor in a more endocrine-resistant population. Patients enrolled in that study would have relapsed on adjuvant endocrine therapy or have had a treatment-free interval of less than a year. Frings said Roche is far more confident in this trial because most of these patients bear ESR1 mutations, for whom oral SERDs are known to work better.
Roche’s first FDA application for giredestrant, bearing a target decision date of Dec. 18, 2026, is in second-line, ESR1-mutated HR+/HER2- breast cancer. The limited pursuit of the ESR1 subgroup was somewhat expected, mirroring past oral SERD approvals; but it was still disappointing given how Roche had touted a benefit in a broad, all-comers population.
“We knew that an [intent-to-treat label] would be an uphill battle, therefore we filed on purpose in ESR1-mutant [patients],” Frings said. But Roche “would leave the door open, depending on the evolution of the data,” to potentially seek a broader label regardless of ESR1 status, especially if more mature OS data are strong, he added.
Eyes on the prize
The FDA has accepted giredestrant for HR+ early-stage breast cancer, Roche announced Monday. After the company spent a priority review voucher, the FDA has set a target decision date of Nov. 30.
At ASCO 2026, a subgroup analysis of the lidERA trial showed the benefit of adjuvant giredestrant regardless of a patient’s menopausal status. Compared with standard endocrine therapy, giredestrant improved invasive disease-free survival (iDFS) by 35% in pre-menopausal women and by 26% in post-menopausal patients.
During the investor call in April, Roche Pharmaceuticals CEO Teresa Graham defended giredestrant’s potential—with a peak sales forecast “well north of” 3 billion Swiss francs—despite the first-line persevERA setback. That’s because the adjuvant setting is a much larger market “with three times more drug-treated patients than in first-line metastatic and a much longer treatment duration [of] around five years,” she said.
However, critics have pointed out that the lidERA trial does not answer how giredestrant fares either against or in combination with CDK4/6 inhibitors, which has recently become a standard treatment for certain patients with early-stage HR+/HER2- breast cancer.
By a cross-trial comparison, giredestrant’s 30% iDFS improvement was comparable to the CDK4/6 drugs, although the trial populations were different. While Verzenio’s and Kisqali’s adjuvant uses—which require combination with endocrine therapy—cover patients at high risk of recurrence, single-agent giredestrant may be an attractive option for low- and medium-risk patients, Frings said.
For patients who want to maximize efficacy, giredestrant could be the best backbone for a combination because it’s the more potent oral SERD, Frings argued.
For now, Roche is conducting a 100-patient substudy in lidERA, evaluating the safety of giredestrant in combination with Eli Lilly’s CDK4/6 drug Verzenio in the adjuvant setting. It also recently started a separate 200-patient study evaluating the safety of giredestrant with Novartis’ Kisqali.
Other companies, including Lilly, AstraZeneca and Menarini, are conducting phase 3 trials to test their oral SERDs in patients who have already received a few years of adjuvant therapy, including a CDK4/6 drug.
Frings argued that this trial design is “convoluted” and “more like a 2-by-2 design” because not all patients will be exposed to a CDK4/6 inhibitor.
Besides, while lidERA showed a treatment effect for the oral SERD early on, that opportunity of early intervention is gone in the competitor design, the Roche exec said. Because those trials’ participants would have previously received two to five years of adjuvant therapy, adherence may be a challenge for continued treatment, too, he added.
As for Roche’s upcoming adjuvant trial, “we must just design it well, get buy-in from authorities,” Frings said. “And then, once we are able to talk about it, we will announce what precisely we will do.”
