Understanding the NATALEE trial—from a methodologist stand of point

NATALEE study presented the primary results of invasive disease-free survival (iDFS) at the 2023 American Society of Clinical Oncology (ASCO) meeting and released its iDFS key subgroup results at the 2023 European Society for Medical Oncology (ESMO) as well as the final iDFS results at the 2023 San Antonio Breast Cancer Symposium (SABCS) before publishing its results at the New England Journal of Medicine (NEJM) in March 2024 (1). Recently, the N0 subgroup data was presented at the just concluded 2024 ASCO meeting (2). Here I would like to discuss the highlights of the methodology design and the common misunderstandings about the results of this pivotal trial.

Compared to the monarchE (3) trial, NATALEE study enrolled a wider population, including patients with anatomical stage II and those without lymph node involvement (N0). Therefore, it is reasonable to use stage II and III as stratification factors in randomization stratification strategy. The baseline characteristics can then be well balanced between the ribociclib group and the control group. At the original protocol, the quota for each stage was 50/50, which allocated sufficient subjects in each subgroup to ensure the stability of the results. Of course, there is room for improvement from a research design point of view. If patients were randomized into three layers: IIA, IIB, and III, more suggestive and reliable conclusions might be obtained when the results in stage II patients was not ideal.

From a protocol amendment perspective, the initial design (4) of this study aimed to enroll 4,000 patients with an estimated hazard ratio (HR) of 0.73. However, after reporting of the negative results of the PALLAS study (5) and the positive results of the monarchE study (3), the independent data monitoring committee (IDMC) re-estimated the sample size according to monarchE data (HR =0.747) and increased the sample size of 1,000 stage III patients and the pre-set number of iDFS events to 500 (stage II patient enrollment was completed at the time of re-estimation). That means 85% power can still be provided even when the final HR is as high as 0.76. Considering the fact the final reported HR of the NATALEE study was around 0.75, this was one of the critical decisions made by the IDMC team that ensured the success of this study and undoubtedly provided valuable lessons for future study design.

In terms of study results interpretation, we noticed that several misunderstandings from clinicians may occur and worth further discussion: (I) some clinical experts challenged the results published in the NEJM that the arms of the iDFS survival curve from two groups crossed around the 40-month mark, thus drawing the conclusion that ribociclib is not beneficial or even inferior to the control group after 40 months. This erroneous conclusion is due to the overlooking of the median follow-up time and number of patients at risk. The median follow-up time for the iDFS events of the intention-to-treat (ITT) population at the NEJM cut-off was 27.7 months, while the crossover occurred at 40 months when the initial patients at risk in each group was just over 10 cases. It is always unreliable to draw any conclusion with limited number of cases. As a matter of fact, this argument was resolved by the prolonged follow-up (33.3 months) results that released at the 2023 SABCS (6), which showed that the crossing of survival curves of iDFS disappeared when the number of patients at risk increased. Therefore, the median follow-up time and the number of patients at risk at each time point must be considered to avoid misleading conclusions when interpreting clinical trial results. (II) Clinicians are accustomed to use time point survival rates (e.g., 3-year survival rates) and median survival time to interpret study results and even for indirect comparisons between different clinical trials in order to conclude “superiority or inferiority”. This is another common misunderstanding because time point survival rates and median survival time cannot comprehensively reflect the overall differences and indirect comparisons between different studies are not appropriate. As another commonly used indicator in clinical trial, HR is relatively more comprehensive and accurate as well as stable compared to time point survival rates and might be considered for indirect comparisons between different studies under certain conditions if necessary (such as NATALEE HR =0.75, monarchE HR =0.75). However, extra cautious should be paid to cross-trial comparisons, especially in clinical trials like NATALEE and monarchE where large differences exist, such as enrolled populations, treatment period, and follow-up time. Nevertheless, such comparisons are generally used to generate study hypotheses. Although not ideal, hypotheses of interest can also be generated through some indirect statistic comparison methods, such as matching adjusted indirect comparison (MAIC) analysis. (III) Another common misunderstanding lies in the interpretation of the results from subgroup analysis. Subgroup analyses are often underpowered due to the limited number of patients and resulting in misunderstanding results. It is inappropriate to conclude whether the result of the subgroup analysis is statistically difference. Neither to exploit the results to compare the efficacy among different subgroup analyses. For example, the 95% confidence interval (CI) of HR for stage III patients in the NATALEE study did not across 1 (statistically significant), while the 95% CI of HR for stage II patients did (not statistically significant), thereby incorrectly concluding the benefit of ribociclib in stage III patients was greater than that in stage II patients. Similar interpretation also occurred in the lymph node negative and positive subgroups. Alternatively, a more reasonable way to interpret the subgroup results is to first consider whether the factors for this subgroup analysis are stratification factors in the randomization process (previously mentioned, ensuring comparability between groups), and then to test whether there is “heterogeneity” (interaction) between groups. And if there was a possible interaction among different subgroups, it needs to be cautiously interpreted from a clinical point of view. In the NATALEE trial, the benefit of ribociclib in term of HR in the anatomical stage II subgroup was very similar to the stage III group. Therefore, above assumption is not appropriate and all available data showing that the benefit of ribociclib group over control group is consistent across all subgroups.

In addition, the results of the N0 subgroup of the NATALEE trial have attracted a lively discussion from many doctors, especially after its recent release at the 2024 ASCO. Regrettably, this study was not stratified according to lymph node status. The number of subjects in the pN0 subgroup was relatively small (~12%) thus it is not possible to obtain reliable conclusions. However, the presented study results from the 2024 ASCO suggested that the baseline characteristics of N0 patients are well balanced between ribociclib group and placebo group which strengthened our confidence in this specific population. Nevertheless, it is possible that there are some unknown confounding factors that may affect the result but have not been identified. Hypothetically, if no discordance of HR was identified, patients in the N0 subgroup who have higher risk of relapse may derive greater iDFS benefit from ribociclib. Further biomarker studies of the NATALEE trial are keenly awaited to better identify those who may benefit the most from this novel treatment. Further discussion is required to achieve a consensus for risk factors and clinical recommendation.

A commonly accepted view of point from clinicians is that overall survival (OS) is the “golden standard” for clinical trials of anticancer treatments due to its unquestionable objectivity and accuracy. However, OS may not always be suitable for diseases that advance slowly and has long-term survival, as longer survival may be more prone to be affected by deaths unrelated to the tumor progression. Therefore, PFS is a well-accepted alternative endpoint in patients with advanced hormone receptor-positive/human epidermal growth factor receptor 2-negative (HR⁺/HER2⁻) breast cancer (7). Considering the fact the survival of HR⁺/HER2⁻ early breast cancer study is usually long and is more likely to be affected by non-tumor related death and other factors, the therapeutic effect in the experimental group may diminish over time as the follow-up period extends. Specific to the OS results of the NATALEE trial, the median OS at 30.4 months follow-up that published at the 2023 ASCO showing a HR result of 0.76 (95% CI: 0.539–1.068) between the ribociclib group and the placebo group. However, by the median follow-up prolonged to 35.9 months, the HR of the updated OS was increased to 0.89 (95% CI: 0.661–1.203). It is important to point out that this change might be related to the small number of events that occurred during the short extended follow-up, and also might be influenced by the deaths from non-tumor causes as well as the effect of subsequent treatments.

Similar alteration of the OS result was also observed in the early follow-ups of the monarchE study which resulted in the changes of Food and Drug Administration (FDA) approval label for abemaciclib (8). Therefore, it is inappropriate to over-interpret the change of OS results at this time when positive results are achieved for the primary endpoint of iDFS. Certainly, as a methodologist, I sincerely hope that NATALEE will achieve positive results in the final OS analysis.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Translational Breast Cancer Research. The article has undergone external peer review.

Peer Review File: Available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-24-32/prf

Funding: None.

Conflicts of Interest: The author has completed the ICMJE uniform disclosure form (available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-24-32/coif). The author has no conflicts of interest to declare.

Ethical Statement: The author is accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Slamon D, Lipatov O, Nowecki Z, et al. Ribociclib plus Endocrine Therapy in Early Breast Cancer. N Engl J Med 2024;390:1080-91. [Crossref] [PubMed]
2. Yardley DA, Untch M, Barrios CH, et al. Baseline (BL) characteristics and efficacy endpoints for patients (pts) with node-negative (N0) HR+/HER2− early breast cancer (EBC): NATALEE trial. J Clin Oncol 2024;42:abstr 512.
3. Johnston SRD, Toi M, O'Shaughnessy J, et al. Abemaciclib plus endocrine therapy for hormone receptor-positive, HER2-negative, node-positive, high-risk early breast cancer (monarchE): results from a preplanned interim analysis of a randomised, open-label, phase 3 trial. Lancet Oncol 2023;24:77-90. [Crossref] [PubMed]
4. Slamon DJ, Fasching PA, Hurvitz S, et al. Rationale and trial design of NATALEE: a Phase III trial of adjuvant ribociclib + endocrine therapy versus endocrine therapy alone in patients with HR+/HER2- early breast cancer. Ther Adv Med Oncol 2023;15:17588359231178125. Erratum in: Ther Adv Med Oncol 2023;15:17588359231201818. [Crossref] [PubMed]
5. Mayer EL, Dueck AC, Martin M, et al. Palbociclib with adjuvant endocrine therapy in early breast cancer (PALLAS): interim analysis of a multicentre, open-label, randomised, phase 3 study. Lancet Oncol 2021;22:212-22. [Crossref] [PubMed]
6. Hortobagyi G, Stroyakovskiy D, Yardley D, et al. Ribociclib (RIB) + nonsteroidal aromatase inhibitor (NSAI) as adjuvant treatment in patients with HR+/HER2− early breast cancer: final invasive disease–free survival (iDFS) analysis from the NATALEE trial. Cancer Res 2024;84:abstr GS03-03.
7. Seidman AD, Bordeleau L, Fehrenbacher L, et al. National Cancer Institute Breast Cancer Steering Committee Working Group Report on Meaningful and Appropriate End Points for Clinical Trials in Metastatic Breast Cancer. J Clin Oncol 2018;36:3259-68. [Crossref] [PubMed]
8. Harbeck N, Rastogi P, Martin M, et al. Adjuvant abemaciclib combined with endocrine therapy for high-risk early breast cancer: updated efficacy and Ki-67 analysis from the monarchE study. Ann Oncol 2021;32:1571-81. [Crossref] [PubMed]