Oncology · Medical Literature
[Comment] Now or later? Switching therapies in metastatic breast cancer
Abstract
In contemporary management of patients with hormone receptor-positive, HER2-negative metastatic breast cancer, it is standard to switch therapies at tumour progression, defined by RECIST criteria in clinical trials, and by other measures (many, ironically, considered non-measurable by RECIST) in clinical practice, including subtler radiological changes, findings on examination, rising tumour markers, or increased symptoms related to cancer.1 Accordingly, in the clinic, a treatment change can be considered for symptomatic progression despite radiologically stable disease, or the same treatment might be continued beyond asymptomatic oligoprogression.
Clinical relevance
In patients with hormone receptor-positive, HER2-negative metastatic breast cancer, should therapy be switched at the time of ctDNA-detected molecular progression (e.g., emergent ESR1 mutation) rather than waiting for conventional radiographic or clinical tumor progression? __SNAPSHOT__ {"studyDesign":"Expert commentary / editorial; not a primary research study. This Comment contextualizes the phase III SERENA-6 randomized controlled trial (double-blind, placebo-controlled), which is the referenced primary evidence.","setting":"International, multicenter academic oncology practice (SERENA-6 trial context); commentary authored from Dana-Farber Cancer Institute, Harvard Medical School, and affiliated institutions","sampleSize":"Not applicable (commentary). Referenced SERENA-6 trial: n=315 randomized (camizestrant + CDK4/6 inhibitor: n=157; aromatase inhibitor + CDK4/6 inhibitor: n=158); screened from 3,256 patients undergoing ctDNA surveillance","population":"Adult patients with histologically confirmed HR-positive, HER2-negative locally advanced or metastatic breast cancer receiving first-line AI plus CDK4/6 inhibitor (≥6 months), with an ESR1 mutation detected on serial ctDNA surveillance in the absence of radiographic or clinical progression","intervention":"Preemptive switch of endocrine backbone from aromatase inhibitor (anastrozole or letrozole) to camizestrant 75 mg orally once daily, with continuation of the same CDK4/6 inhibitor (palbociclib, ribociclib, or abemaciclib), guided by ctDNA-detected emergent ESR1 mutation before clinical or radiographic progression","comparator":"Continuation of aromatase inhibitor plus CDK4/6 inhibitor (conventional therapy-switch-at-progression paradigm) with camizestrant placebo","primaryOutcome":"Progression-free survival (PFS) as assessed by investigator (SERENA-6 primary endpoint)","followUp":"Median 23.5 months (IQR 17.9–32.1) at extended analysis (data cutoff January 2, 2026); median 12.6 months at first interim analysis","fundingSource":"Industry (AstraZeneca funded SERENA-6; commentary authors' funding not reported in abstract)"}
Key findings
This commentary discusses the central finding of SERENA-6: switching from aromatase inhibitor to camizestrant upon ctDNA detection of an emergent ESR1 mutation — before radiographic or clinical progression — significantly prolonged progression-free survival. At the first interim analysis (median follow-up 12.6 months), median PFS was 16.0 months (95% CI 12.7–18.2) in the camizestrant arm versus 9.2 months (95% CI 7.2–9.5) in the AI continuation arm (HR 0.44; 95% CI 0.31–0.60; P<0.0001). At extended analysis (median follow-up 23.5 months), updated median PFS was 16.8 months (95% CI 14.7–19.4) versus 9.2 months (95% CI 7.2–9.7), (HR 0.45; 95% CI 0.34–0.59; nominal P<0.0001) — representing a 55% relative reduction in risk of progression or death. The absolute PFS gain was approximately 7.6 months (calculated). The 12-month PFS rate was 60.7% versus 33.4%, and the 24-month PFS rate was 29.7% versus 5.4%. PFS2 (time to second progression) also favored camizestrant (median gain 6.6 months; HR 0.63; not further specified in abstract). Time to deterioration in patient-reported global health status/QoL was 21.0 months versus 6.4 months (HR 0.54; 95% CI 0.34–0.84). Chemotherapy/ADC-free survival was 22.6 versus 18.7 months (HR 0.64). Overall survival data remain immature (HR 0.87; 95% CI 0.57–1.30; 30% maturity); no significant OS benefit has been demonstrated. Adverse events: grade ≥3 AEs were more frequent with camizestrant plus CDK4/6 inhibitor (60.0% vs. 45.8%), predominantly CDK4/6 inhibitor–associated hematologic toxicities. Photopsia occurred more frequently in the camizestrant arm (20.0% vs. 7.7%) but was predominantly low-grade with no associated structural ocular changes. Discontinuation due to adverse events was low in both arms (1.3% vs. 1.9%). The commentary also highlights the tension between clinical progression definitions: RECIST criteria govern trial endpoints, while clinical practice involves subtler signals including rising tumor markers, exam findings, and symptom changes — underscoring that the question of 'when to switch' is clinically complex.
Practice implications
SERENA-6 — the study this commentary directly addresses — establishes proof of concept that a ctDNA-guided, pre-progression therapy switch (at the time of ESR1 mutation emergence) significantly prolongs PFS compared to the conventional standard of switching at clinical or radiographic progression in HR+/HER2- advanced breast cancer. This represents the first global registrational Phase III trial to demonstrate the utility of prospective ctDNA monitoring to direct a preemptive endocrine therapy change. However, several important limitations must be acknowledged for clinical interpretation: (1) Overall survival benefit has not been demonstrated; OS data are immature at 30% maturity and the confidence interval crosses 1.0, making OS conclusions premature; (2) The FDA's Oncologic Drugs Advisory Committee (ODAC) voted 6–3 that the data did not demonstrate clinically meaningful benefit for this indication, citing uncertainty about whether PFS and PFS2 are valid surrogates when the treatment change occurs at molecular rather than clinical progression — this regulatory concern is critical; (3) No crossover was permitted from the AI arm to camizestrant at radiographic progression, limiting interpretation of whether 'now' versus 'later' switching ultimately changes OS; (4) Camizestrant remains investigational and is not currently FDA-approved for this indication. The commentary by Valenza, Regan, and Burstein contextualizes this evidence within the nuanced reality of clinical practice, where progression is defined by signals beyond RECIST — including rising tumor markers, clinical symptoms, and subtle imaging changes. It is only available as an abstract; the full text may contain additional context and nuance not captured here. Findings are most applicable to patients on first-line AI + CDK4/6i who are undergoing serial ctDNA surveillance, but applicability to patients not in active ctDNA monitoring programs requires infrastructure not universally available. __LIMITATIONS__ 1. This is a commentary, not a primary study; clinical data are derived solely from SERENA-6 as referenced. Only the abstract was provided, limiting full critical appraisal. 2. OS is not yet mature and the interim OS HR (0.87; 95% CI 0.57–1.30) does not exclude harm or equivalence; OS remains the definitive efficacy endpoint. 3. The FDA ODAC voted 6–3 against demonstrating clinically meaningful benefit, specifically questioning whether PFS2 is a valid surrogate when the experimental arm initiates treatment change at molecular rather than clinical progression — a fundamental methodological concern. 4. No crossover to camizestrant was permitted in the AI continuation arm upon radiographic progression, making it impossible to distinguish whether benefit is attributable to the drug itself, the timing of the switch, or both. 5. PFS2 as a regulatory endpoint was deemed potentially unacceptable by the FDA in this context, undermining the surrogate validity argument. 6. The study enrolled a highly selected population (ESR1-positive on ctDNA, no progression) from a screened pool of >3,000 patients; generalizability to centers without established serial ctDNA surveillance programs is uncertain. 7. Industry funding (AstraZeneca) introduces potential bias in study design, data analysis, and dissemination. 8. PFS improvement, while statistically significant and numerically substantial, has not translated into confirmed OS improvement; the clinical meaningfulness of PFS as the primary endpoint in this context is actively debated. 9. The higher grade ≥3 AE rate (60.0% vs. 45.8%) and photopsia burden in the camizestrant arm require longer-term safety follow-up. 10. Most enrolled patients were White (63%); racial and ethnic diversity was limited, potentially affecting generalizability. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ SERENA-6 demonstrates that preemptive switching from an aromatase inhibitor to camizestrant upon ctDNA-detected ESR1 mutation emergence — before clinical or radiographic progression — significantly prolongs PFS (16.8 vs. 9.2 months; HR 0.45) and delays QoL deterioration in HR+/HER2- advanced breast cancer, but OS benefit has not been established and the FDA ODAC voted against clinically meaningful benefit, leaving regulatory approval uncertain. This ctDNA-guided strategy is scientifically compelling and conceptually practice-informing, but should not be adopted as a new standard of care until regulatory clarity is achieved and OS data mature. Clinicians should follow regulatory updates closely, as the question of 'when to switch' remains clinically and statistically unresolved.