A single institution experience with palbociclib toxicity requiring dose modifications

Palbociclib was initially approved on the basis of a small number of patients treated on the PALOMA 1 study. The PALOMA 2 and PALOMA 3 trials included over 700 patients and confirmed grade 3 and 4 neutropenia as the most common side effect from palbociclib and endocrine therapy [2, 3]. Patients who participate in registration trails are generally stringently selected for performance status, age, and limited prior therapies based on prespecified inclusion criteria [5]. A non-clinical trial experience with new agents is important as drug toxicities in routine clinical practice may be more common or more severe than those observed in clinical trial settings. The incidence of grade 3 and 4 neutropenia in our population (54.8%) was comparable to the incidence reported in the aforementioned studies. Although we recognize that our study was conducted through a specialized breast medical oncology clinic and may not be entirely representative of community oncology practices, we also observed the use of non-standard dose modification schemes for palbociclib that are reflective of the variability in real-world individual provider practice patterns. For example, the package insert recommends providers to hold palbociclib for grade 3 neutropenia on day 1 of each cycle, check a CBC in 1 week, and resume at same dose for next cycle if recovered to grade ≤ 2; dose reductions to next lower dose are recommended for grade 3 neutropenia > 1 week, recurrent grade 3 neutropenia, grade 3 neutropenia with fever and/or infection, or grade 4 neutropenia [4]. For grade 1–2 hematologic and non-hematologic toxicities, no dose modifications are required. For grade ≥ 3 non-hematologic toxicities, it is recommended to hold palbociclib until symptoms resolve to grade 1–2 and resume at next lower dose. In our study, we saw dose modifications for grade 1–2 neutropenia (4.8%), grade 1–2 thrombocytopenia (9.4%), grade 2 fatigue (2.4%), grade 2 mucositis (7.1%), grade 1 renal injury (2.4%), and for anticipated tolerance due to age (4.7%). Of the six patients (15.8%) managed by holding palbociclib and resuming by next cycle or holding indefinitely, two of these patients ultimately required dose adjustments. We also observed modifications by way of every other day dosing and direct dose reduction to 75 mg daily (Table 2) that altogether underscore variability in management practices per provider discretion.

Of note, another CDK4/6 inhibitor, ribociclib, was recently approved in the first-line treatment of postmenopausal, HR-positive/HER2-negative advanced breast cancer based on superior median PFS with ribociclib + letrozole (not reached) compared to placebo + letrozole (median PFS 14.7 months, hazard ratio 0.56 (95% CI 0.43–0.72, p < 0.0001) in the phase III MONALEESA-2 trial [6]. Also recently approved was the CDK4/6 inhibitor, abemaciclib, as monotherapy or in combination with fulvestrant in the treatment of HR-positive, HER2-negative advanced or metastatic breast cancer with disease progression following endocrine therapy and, if applicable, prior chemotherapy based on the phase II MONARCH 1 and phase III MONARCH 2 trials [7, 8]. MONARCH 1 was an open-label, single-arm trial investigating single-agent abemaciclib 200 mg oral twice daily until disease progression and showed an ORR of 19.7% (95% CI 13.3–27.5), which was its primary objective [7]. Single-agent abemaciclib was associated with a grade 3–4 neutropenia rate of 26.8%. The randomized, placebo-controlled phase III MONARCH 2 investigated abemaciclib 150 mg oral twice daily with fulvestrant (standard dosing) versus fulvestrant alone and showed a superior median PFS of 16.4 months in the combination arm versus 9.3 months in the fulvestrant alone arm (HR 0.553, 95% CI 0.449–0.681, p < 0.0000001) [8]. All-grade treatment-emergent neutropenia was more common in the combination arm vs. fulvestrant alone (46.0% vs. 4.0%) though diarrhea was the most frequent AE (86.4% vs. 24.7%). The rate of grade 3–4 neutropenia in the ribociclib arm was 59.3%, which is comparable to the 54–66.4% grade 3–4 neutropenia rate observed in the palbociclib arms of the PALOMA studies [1‐3]. Further studies are warranted to validate if our data are generalizable to other CDK4/6 inhibitors in routine clinical practice.

On analysis of our age ≥ 65 subgroup (n = 31), we found comparable rates of initial dose modifications (38.7%) and subsequent dose changes (25.0%) to the overall cohort. Similarly, most dose modifications occurred early in the treatment course (91.7% during cycle 1) and were most frequently a result of grade 3–4 neutropenia (46.6%), grade 2–3 mucositis (13.3%), or anticipated tolerance due to age (13.3%). Of the 12 patients that required dose modifications in palbociclib, the most common method was reduction to 100 mg (50.0%) followed by reduction to 75 mg (33.3%). This is consistent with findings from a recent pooled analysis of PALOMA trials that included 221 patients aged ≥ 65 to 74 and 83 patients aged ≥ 75 out of 872 patients treated with palbociclib + letrozole or fulvestrant [9]. They showed that the incidence of AEs and palbociclib discontinuations were similar in the overall cohort and in patients aged ≥ 65 to 74 and ≥ 75. The incidence of all-grade and grade 3–4 neutropenia was also similar across all age groups. Furthermore, improvements in efficacy endpoints were seen across all age groups without any clinically relevant differences in pharmacokinetics. Our results support their conclusions that there was no difference in tolerance of palbociclib + endocrine therapy between patients aged ≥ 65 and other patients with HR-positive, HER2-negative advanced breast cancer; and a dose adjustment based on age is not required.

It is anticipated that new side effects will be identified in the postmarketing period of palbociclib. It is estimated that half of all serious AEs are only completely realized more than seven years after FDA approval [10]. The incidence of stomatitis was 11–15% in the PALOMA studies with < 1% grade 3–4 [1, 3, 11]. The manifestations of mucositis/stomatitis in our patients varied, but typically presented as tiny mucosal lesions that did not interfere with oral intake and spontaneously resolved. Absence of overt ulcerations, but rather complaints of an inflamed, sore mouth were also seen. More severe cases involved swelling of the tongue and lips and impairment in the ability to consume food. The occurrence varied in onset as early as cycle 1 to a relatively protracted course beyond cycle 5 requiring subsequent dose reductions and occurred often in the setting of grade ≥ 2 neutropenia. We also described a unique case of palbociclib-related pneumonitis observed in a 72-year-old patient who was started on second-line palbociclib (125 mg) and fulvestrant. Along with supportive care, her pneumonitis resolved within 3 months of discontinuation of palbociclib. To our knowledge, this is the first case described in the literature of palbociclib-associated pneumonitis. The identification of grade 2–3 mucositis comprising 9.5% of AEs leading to dose modifications in our cohort also highlights a poorly characterized but relevant AE associated with palbociclib. Mucositis is now recognized as a side effect of palbociclib and has been added to the package insert.

In conclusion, the addition of palbociclib to endocrine therapy in advanced HR-positive, HER2-negative breast cancer results in dose modifications in 38.0% of patients due to AEs with 18.4% requiring subsequent dose changes. Most palbociclib dose modifications occurred during the first two cycles. The most common reason for palbociclib dose modifications was grade 3–4 neutropenia (54.8%). Most providers (65.8%) dose-reduced palbociclib from 125 mg to 100 mg as their preferred method of dose modification when needed. Grade 2–3 mucositis comprised 9.5% of AEs leading to dose modifications, and we described the first case of palbociclib-related pneumonitis; both toxicities are in need of further recognition and characterization. Our findings provide context outside of a clinical trial that can inform ongoing studies of larger size and prospective design evaluating the safety and feasibility of palbociclib-based therapies. This study is limited by its retrospective design. However, our data provides useful insights into the prescribing patterns of physicians while they become comfortable with use of a new agent.