Here we present data on ER+ HER2- MBC patient treated with ribociclib and endocrine therapy from three Danish departments of oncology. First and foremost, our analyses indicate that patients who receives dose reductions during ribociclib treatment do not have a shorter PFS compared to patients who continue receiving ribociclib in full dosage. Even when adjusting for patients with late dose reduction, there is a tendency (p=0.078) towards better effect in patients that have been dose reduced. This tendency could be due to unknown confounding given the chosen study design or remaining immortal time bias. However our results indicate that dose reduction of ribociclib is safe, and the results fits preliminary results from MONALEESA trials subgroup analysis [27]. Similar results have previously been published concerning the CDK4/6 inhibitor Palbociclib [28]. Still, caution is advised in concluding that ribociclib in lower doses (200-400mg) have equal (or better) efficacy than full dosage (600mg). For this, more real-world experience and research are needed, including randomized clinical trials (e.g. ClinicalTrials.gov #: NCT03822468).
Overall, our study population consisting of both post-/premenopausal women treated with ribociclib and an AI or fulvestrant should represent patients found within all three MONALEESA trials[15–17]. Unlike MONALEESA-2 where all patients were treated in a first-line setting, here 25.0% of patients received ribociclib treatment as second line or beyond. With that in mind, our median PFS of 19.2 months (95%-CI:14.3-not reached) seems slightly lower than the 25.3 months (95%-CI = 23.0-30.3) observed in MONALEESA-2[15], but closer to results from MONALEESA-3. Here the median PFS was 20.5 months (CI-95% = 18.5-23.5) and 22.7% of included patients had received up to one line of endocrine therapy for advanced disease [16]. Findings from both MONALEESA 3 and 7 suggest that patients derive beneficial effect of ribociclib despite previously having received up to one line of endocrine therapy [16] or chemotherapy[17]. However, in other retrospective studies on CDK 4/6 inhibitors, actual median PFS is, unsurprisingly, lowered in patients receiving ribociclib treatment beyond a first-line setting[12, 29].
Another consideration is that, in contrary to the MONALEESA trials, our real-world study population also include a small proportion (7%) of patients with performance score of 2. Poor performance score might negatively impact survival in breast cancer patients[30, 31]. Overall, comparison of PFS from clinical trials to real-world evidence should be done with caution due to differences in clinical settings and potential unknown differences in patient characteristics. However, given our inclusion of patients receiving ribociclib in a second line or beyond and few patients with a performance score of 2, we report an efficacy of ribociclib in ER+ HER2- MBC patients comparable to results from the MONALEESA trials.
We identified a group of 20 patients who only received ribociclib between a few days and just merely completing the single treatment cycle (28 days). We found no specific baseline characteristics associated with this group of patients (Table 1). Underlining the fact that it can be difficult to determine whether patients should initiate ribociclib treatment even when they are formally eligible for the treatment.
In the present study, 46.9% of patients required at least one dose reduction, this seems comparable to results from MONALEESA-2 (57.5%), MONALEESA-3 (37.9%) and MONALEESA-7 (35%). We found lower risk of requiring dose reduction, when the endocrine drug co-administrated with ribociclib was fulvestrant compared to an aromatase inhibitor (ORa = 0.30, CI-95%: 0.12-0.73). While this difference could be explained by a different safety profile between fulvestrant, generally aromatase inhibitors are thought to have a similar safety profile as fulvestrant[32, 33] and both MONALEESA-2 and 3 report almost same frequency of adverse events[15, 16]. While our regression analysis indicates no association between previous treatment for advanced disease and dose reduction, a combination of the patient’s history of both previous (neo)adjuvant treatment and previous lines of endocrine/chemotherapy could be a possible explanation. Moreover, the patients with lymph node involvement at baseline have reduced odds of requiring dose reduction. In the tolerability data, there is no difference between baseline characteristics or adverse events reported in patients with lymph node involvement compared to those without lymph node involvement (data not shown). So, it remains speculative, what causes this difference in risk.
Generally, the commonly reported adverse events in the present study supports a manageable safety profile as evident from the MONALEESA trials[15–17]. Occurrence of neutropenia grade III (45.3%) and IV (7.0%) in this real-world population, seems comparable to numbers observed across the MONALEESA trials, with grade III occurring between 46.6% to 52.4% and similarly grade IV in 6.8% to 10%. During treatment three (2.3%) patients developed postbaseline QTc prolongation requiring dose reduction or discontinuation. In addition, three patients were non-eligible for ribociclib treatment due to QTc prolongation at ECG control 14 days posttreatment start. Compared to the MONALEESA trials where 3.6-7% developed postbaseline QTc prolongation of >480ms[15–17], our real-world data supports recommendations that physician should monitor ECG throughout a ribociclib treatment course.
Some of the study limitations naturally adhere to the chosen study design. Retrospective studies rely on precise information from electronic health records and lack of randomization means results could be impacted by confounding. Despite inclusion of patients from three different institutions, our moderate sample size (n=128) should be considered before generalizing our results into broader populations. Furthermore, as collection and categorizing of adverse events were done retrospective by the authors, there is a risk of misclassification. This risk is not present concerning neutropenia, thrombocytopenia and ALAT increase, as these adverse events were assessed by blood samples available for all patients (except one).Finally, our PFS analysis of patients receiving ribociclib in reduced dose, does not differentiate if patient’s final dose was 400mg or 200mg.