Results

Sociodemographic and clinical characteristics

A total of 39 patients were enrolled, with a median age of 58 years (IQR 52.0–67.0) (Table 1). Most (94.9%) had metastatic breast cancer, with metastasis in visceral organs (n = 16), bone (n = 19), other locations (n = 13), and brain (n = 1), while 2 patients (5.1%) had locally advanced disease. At abemaciclib initiation, the median time since diagnosis of metastatic disease was 2.0 months. Patients registered in the PSP a mean of 7.7 days after the start of abemaciclib treatment, and most (94.9%) started at 150 mg twice daily (BID) dose. The median follow-up time was 5.6 months.

Characteristics | Value
Age, median years (IQR) | 58.00 (52.0–67.0)
BMI, mean (SD) | 27.20 (4.95)
Normal (18.5 ≤ BMI < 25), n (%) | 17 (43.59)
Overweight (25 ≤ BMI < 30), n (%) | 12 (30.77)
With obesity (BMI ≥ 30), n (%) | 10 (25.64)
Carcinoma type, n (%)
Locally advanced BC | 2 (5.13)
Metastatic BC* | 37 (94.87)
Visceral | 16 (41.03)
Bone | 19 (48.72)
Brain | 1 (2.56)
Other | 13 (33.33)
Median time since diagnosis, months (IQR) | 2.00 (1.0–26.0)
Median follow-up time, months (IQR) | 5.59 (5.19–5.78)
Median time from start of abemaciclib to PSP registration, days (IQR) | 7.00 (0.0–14.0)
Comorbidities, n (%)
Yes | 19 (48.72)
No | 20 (51.28)
Abemaciclib starting dose, n (%)
50 mg BID | 1 (2.56)
100 mg BID | 1 (2.56)
150 mg BID | 37 (94.87)
Bowel movements (prior to start abemaciclib treatment). Number of daily bowel movements, n (%)**
0 | 6 (18.18)
1 | 24 (72.73)
2 | 3 (9.09)
Type (according to the Bristol stool scale), n (%)***
Type 1 | 1 (3.57)
Type 2 | 4 (14.29)
Type 3 | 9 (32.14)
Type 4 | 11 (39.29)
Type 5 | 1 (3.57)
Type 6 | 1 (3.57)
Type 7 | 1 (3.57)

Frequency and types of stools

Before starting abemaciclib, most participants reported regular bowel habits according to the Bristol scale: 72.7% had one daily bowel movement and 75% had normal stool types (3, 4, or 5) (Table 1).

Overall, 35 of the 39 participants (89.7%) experienced diarrhea during the study, including three patients (7.7%) with grade 3 events. No patients experienced grade 4–5 diarrhea. The median time to first diarrhea episode was 10 days (IQR 3–21) after the initiation of abemaciclib. All events were attributed to abemaciclib. The proportion of patients with diarrhea decreased from 62 to 50% from weeks 2 to 24, respectively (Table 2).

 | Number of patients
Diarrhea | 2 weeks (n = 39) | 4 weeks (n = 38) | 8 weeks (n = 38) | 12 weeks (n = 38) | 24 weeks (n = 36) | Total (n = 39)
Yes, n (%) | 24 (61.54) | 27 (71.05) | 25 (65.79) | 21 (55.26) | 18 (50) | 35 (89.74)
No, n (%) | 15 (38.46) | 11 (28.95) | 13 (34.21) | 17 (44.74) | 18 (50) | 4 (10.26)
MD | 0 | 1 | 0 | 0 | 1 | 
The reported n refers to patients continuing in the study. One patient discontinued at week 4, one at week 12, and four at week 24; therefore, n is adjusted accordingly for follow-up visits
Most severe grade per patient, n (%) (n = 39)
CTCAE
Grade 1 | 18 (46.15) | 
Grade 2 | 14 (35.90)
Grade 3 | 3 (7.69)
Grade 4 | 0 (0)
Grade 5 | 0 (0)
No diarrhea | 4 | 

Treatment modifications due to diarrhea

No patient discontinued abemaciclib due to diarrhea. Among the 35 patients with diarrhea, 9 (23.1%) modified their treatment, either reducing the dose and/or temporarily interrupting treatment on one or more occasions (Supplementary Fig. 2). These nine patients accounted for nine dose reductions and five temporary interruptions. Eight patients reduced the dose once (all from 150 mg BID to 100 mg BID), and one patient did so twice (eventually reducing to 50 mg BID). Three patients had temporary interruptions prior to dose reduction, and one had two interruptions without any dose change (Fig. 1).

Fig. 1: Diarrhea-related modifications to abemaciclib treatment

In four cases, diarrhea was the sole reason for the dose reduction; in another four, it was combined with other unspecified causes. The median time to dose reduction was 52.5 days (IQR 28.0–83.5). Reductions were associated to grade 1 (_n _= 6) or grade 2 (n = 3) episodes (Fig. 1). None of the three patients with grade 3 diarrhea had a dose modification. Figure 2 shows the Kaplan–Meier curve for time to dose reduction due to diarrhea.

Fig. 2: Kaplan–Meier curve of time to first dose reduction due to diarrhea

Four patients (10.3%) temporarily interrupted abemaciclib due to diarrhea (Fig. 1), one of whom had two interruptions. Interruptions were associated with grade 1 (_n _= 2) or grade 2 (n = 3) episodes (Fig. 1). The median time to first interruption was 34.5 days (IQR 27.0–44.0), with a median duration of 11.0 days (IQR 5.0–15.0).

Patients’ self-care and diarrhea management

Loperamide prescription was the most common strategy for managing diarrhea, reported by over 75% of patients between weeks 2 and 12, with use declining by week 24. Other frequent strategies included increasing fluid intake, astringent diet, and food restrictions—similarly recommended by both the HCPs and the PSP (Supplementary Table 3). Over 60% of the patients found each of these strategies helpful with loperamide being rated as helpful by 100% of patients at week 12.

Patients reported feeling confident and supported in managing diarrhea and their disease. By week 24, over 70% agreed or strongly agreed that the PSP helped them manage diarrhea, adhere to treatment and feel heard and supported (Supplementary Fig. 3).

Adherence to treatment (SMAQ)

Over 80% of patients were classified as adherent, with adherence highest at week 2 and lowest at week 24 (Table 3).

Time (weeks) | 2 weeks | 4 weeks | 8 weeks | 12 weeks | 24 weeks
Adherent, n (%) | 32 (91.4) | 29 (87.9) | 28 (84.8) | 29 (90.6) | 24 (80.0)
Non-adherent, n (%) | 3 (8.6) | 4 (12.1) | 5 (15.1) | 3 (9.4) | 6 (20.0)
Total * | 35 (100) | 33 (100) | 33 (100) | 32 (100) | 30 (100)
MD | 4 | 6 | 5 | 6 | 7

HRQoL assessment

Mean FACIT-D scores ranged from a low of 104.6 (IQR: 78.0–125.0) at week 4 to a high of 112.8 (IQR: 98.0–132.5) at week 24, on a scale from 0 to 152. Scores declined at weeks 4 and 8 and returned to baseline levels by weeks 12 and 24 (Fig. 3A). Diarrhea subscale scores followed a similar trend, peaking at week 2 (36.4, IQR: 34–42), declining to their lowest at week 8 (33.3, IQR: 25–41) and recovering by the end of follow-up (Fig. 3B).

Fig. 3: Quality of life of patients (FACIT scores). Panel A shows the total FACIT-D score, and panel B shows the FACIT-D Diarrhea Subscale during follow-up

Evaluation of patient persistence, discussions, and satisfaction with the PSP

All patients who continued abemaciclib remained enrolled in the PSP. Two patients discontinued both abemaciclib and the PSP for personal reasons, not reported associated with diarrhea, while two others left the PSP after stopping abemaciclib for reasons other than diarrhea, including one death.

Almost all patients (94.9%) contacted the PSP at least once, discussing diarrhea and nutritional issues in similar proportions (Supplementary Fig. 4).

By week 24, over 70% of the patients reported being satisfied or completely satisfied with the PSP across all evaluated aspects, including information on disease, side effects, diarrhea management, and emotional support (Supplementary Fig. 5 and Supplementary Table 4).