Oncology development
Why Overall Survival Takes Years to Measure in Early Breast Cancer
Explore late recurrence in HR-positive, HER2-negative early breast cancer, survival findings from monarchE and NATALEE, and what oncology trial teams can do to protect long-term follow-up.
By Lampgen Clinical Research

4 min read · estimated
What the source reported
In hormone receptor–positive, HER2-negative early breast cancer, a trial can show a reduction in recurrence years before it can determine whether people live longer. Recurrences can happen well after initial treatment, and it takes time for enough deaths to occur for a planned overall survival analysis. Stephen Johnston’s September 2026 Lilly Medical discussion traces that lag through the history of adjuvant endocrine therapy, which is given after surgery to lower recurrence risk.
Endocrine therapy had survival evidence long before the recent CDK4/6 inhibitor studies. In ATLAS, continuing tamoxifen to 10 years lowered overall mortality compared with stopping at five years among women with estrogen receptor–positive disease. The reduction in breast cancer mortality was more apparent after year 10. A separate Early Breast Cancer Trialists’ Collaborative Group analysis pooled data on 62,923 women who were disease-free at year five and scheduled to stop endocrine therapy. Recurrences continued through years 5–20 after diagnosis, with distant recurrence risk varying substantially by initial tumor size and nodal involvement.
Invasive disease-free survival combines invasive recurrence, certain new invasive cancers and death. Overall survival measures time from randomization to death from any cause. Treatment after recurrence and deaths from other causes can affect the survival comparison. That is why an earlier recurrence-related result and a later survival analysis can answer different questions about the same regimen.
The phase 3 monarchE trial randomized 5,637 patients with node-positive, high-risk HR-positive, HER2-negative early breast cancer to standard endocrine therapy with or without two years of abemaciclib. At 76.2 months of median follow-up, the planned analysis of overall survival, a key secondary endpoint, met its prespecified significance boundary (hazard ratio 0.842, 95% CI 0.722–0.981, P = .027). Estimated seven-year survival was 86.8% versus 85.0%. Seven-year invasive disease-free survival was 77.4% versus 70.9%, a 6.5-percentage-point difference on that separate endpoint.
The October 2026 NATALEE update adds another example. The phase 3 trial enrolled 5,101 adults with eligible stage II or III disease, including some with high-risk node-negative disease. At a median survival follow-up of 66.5 months, three years of ribociclib plus a nonsteroidal aromatase inhibitor was associated with a 2.1-percentage-point higher estimated six-year survival than the aromatase inhibitor alone (hazard ratio 0.809, 95% CI 0.661–0.989). This was a prespecified descriptive analysis with a nominal one-sided P value. The studies enrolled different populations, used different regimens and assessed survival at different time points. Each result informs decisions for the population and regimen it studied.
For eligible patients, mature data make it possible to discuss absolute survival differences alongside baseline recurrence risk and treatment burden. The U.S. labels describe different eligibility and monitoring requirements. Abemaciclib can cause diarrhea and neutropenia. Ribociclib also carries risks including neutropenia, liver toxicity and QT prolongation, with corresponding lab and ECG monitoring. The NATALEE update reported no new safety signals at six years.
What this means for trial teams
Plan follow-up before enrollment. Set out the protocol’s schedule for collecting survival status, recurrence and subsequent anticancer treatment. Specify which site team member owns each contact and who takes over when a coordinator leaves, a site winds down or a participant transfers care. Carry that workload into the staffing and budget plan for the years after dosing ends.
Make continued participation feasible. Explain how stopping study treatment differs from withdrawing consent for further follow-up. Work within each participant’s consent, the protocol and local requirements. Where allowed, use phone follow-up, record review or coordination with a local oncology practice to reduce avoidable visits. Ask participants which contact method and language support work for them.
Review completeness before the data cut. Track overdue contacts, time since last confirmed vital status and unresolved recurrence records. Review patterns by site and region, and by participant subgroup where appropriate and permitted. An acceptable study-wide completion rate can conceal persistent gaps at community sites or among people who face travel and language barriers.
An oncology CRA can verify critical dates against the available source records and escalate missing or conflicting information. A clinical trial manager can coordinate resolution across sites, the CRO and data management and plan coverage through the final analysis. Lampgen supports sponsors and CROs with oncology CRA staffing and clinical operations resourcing for that work.
Sources
- Lilly Medical, Stephen Johnston on overall survival in the adjuvant setting, September 30, 2026
- Davies et al., ATLAS randomized trial, The Lancet, 2013
- Pan et al., 20-year recurrence risk after five years of endocrine therapy, NEJM, 2017
- Johnston et al., monarchE overall survival analysis, Annals of Oncology, 2026
- Slamon et al., NATALEE six-year analysis, Annals of Oncology, October 2026
- FDA, clinical trial endpoints for cancer drugs and biologics
- FDA, ICH E6(R3) Good Clinical Practice
- DailyMed, Verzenio U.S. prescribing information
- DailyMed, Kisqali U.S. prescribing information
- Lampgen Clinical Insights, participation burden and patient access
- Lampgen Clinical Insights, when illness interrupts a trial visit