Page 145 - 2022_01-Haematologica-web

Impact of discontinuation/modifications on outcome
Median PFS and median OS for early discontinuation due to any reason except PD was 24.3 months (95% CI: 20.8–31.9) and not reached (NR; 95% CI: 24.3–NR), respectively, compared with 52.3 (95% CI: 47.9–NR) and NR for all patients in the VenR arm and NR (95% CI: 52.3–NR) and NR in patients who completed venetoclax treatment. Discontinuing treatment early (for any reason except PD) was significantly associated with shorter PFS (n=181; HR 5.98, 95% CI: 3.31–10.82; P<0.0001). A simi- lar adverse impact on PFS was observed for patients who discontinued treatment due to an AE (n=174; HR 5.82, 95% CI: 2.93–11.57; P<0.0001); this inferior survival was consistent with treatment discontinuation due to veneto- clax-related AE (n=174; HR 2.34, 95% CI: 1.10–4.98; P=0.0272). For OS, modeling was insufficiently powered, with too few deaths (six of 140, 4.3%) in the group of patients completing 2 years of venetoclax therapy, hence it was not performed.
Overall, the median duration of venetoclax therapy for patients who discontinued venetoclax early (n=54) was 11.3 months (range, 0.1–24.9; Table 3), compared with 24.4 months (range, 0.1–27.9) for all patients. In patients discontinuing due to AE, median treatment duration was 11.3 months (range, 0.5–23.3), compared with 16.4 months (range, 4.4–24.9) for patients discontinuing for PD (P=0.063, Table 3). The majority of discontinuations occurred after completion of the combination therapy (VenR) phase, with 15 patients discontinuing during the combination therapy period (cycles 1–6, 0 to <6 months; one patient was determined as discontinued during post- combination/follow-up but is included in the 0 to <6 months group in Table 3), nine of these due to AE.
Table 2. MURANO: cause of death for patients who discontinued treatment.
Discontinuation due to AE was more common earlier (34.5% [ten of 29] between 0 and <6 months vs. 20.7% [six of 29] for ≥18 months), with a visible trend for fewer discontinuations due to AE as time progressed (Table 3).
Overall, 41 AE led to treatment discontinuation in 32 patients (16.5%; Table 4); however, the primary reason for early venetoclax discontinuation was only determined as AE for 29 patients (primary reasons for the other three patients were physician decision [n=2] and death [n=1]). Notably, 22 of 29 (75.9%) patients had a dose reduction or interrupted treatment before prematurely discontinu- ing venetoclax due to an AE and 18 of 29 (62.1%) patients discontinued treatment early due to venetoclax related AE (Online Supplementary Table S2). The most common classes of AE (MedDRA preferred terms) leading to dis- continuation were blood and lymphatic system disorders (7.2%); neoplasms benign, malignant and unspecified (2.6%); general disorders and administration site condi- tions (2.1%); and infections and infestations (2.1%). The most common individual AE resulting in treatment dis- continuation were neutropenia (six of 194; 3.1%) and thrombocytopenia (five of 194; 2.6%).
Objective response rate (ORR: complete response [CR], CR with incomplete marrow recovery, partial response [PR], and nodular PR) for patients who prematurely dis- continued venetoclax due to AE was 79.3% (23 of 29; Table 4). Best minimal residual disease (MRD) response in peripheral blood (measured by allele-specific oligonu- cleotide polymerase chain reaction) was undetectable MRD (uMRD; where the threshold for MRD was defined as one tumor cell per 104 white cells) for 62.1% of these patients (18 of 29), while the uMRD rate at the end of combination treatment response visit was 48.3% (14 of 29; Table 5). Patients who had an uMRD status on or before venetoclax discontinuation and did not have PD before venetoclax discontinuation (n=24), had a median PFS calculated from cessation of 26.6 (95% CI: 4.7–32.4) months. There were no robustly significant baseline demographic predictors for discontinuation due to AE; baseline covariates were not significantly associated with discontinuation due to venetoclax related AE, but dura- tion of response to most recent prior therapy was margin- ally significant (P=0.0467) for treatment discontinuation due to non-related AE (Online Supplementary Table S3).
Table 3. MURANO: discontinuation of venetoclax treatment for all patients, due to adverse events and due to progressive disease.
Cause of venetoclax discontinuation
Adverse event
Cause of death
Acute respiratory failure
Cardiac failure
Colorectal cancer
Respiratory failure
Sepsis secondary to transformed acute myeloid leukemia
Metastatic malignant melanoma
Myelodysplastic syndrome (related to venetoclax)
Myocardial infarction (related to venetoclax) Pancreatic carcinoma
Pneumonia
Status epilepticus
Sudden death
Progressive disease Pneumonia
Pneumonia
Sudden cardiac death
Progressive disease
Sepsis (related to venetoclax)
Not reported (patient had gastric cancer and was transferred to palliative care)
Patients who discontinued venetoclax
and died (n=22)
1 (4.5%) 1 (4.5%) 2 (9.1%) 1 (4.5%)
1 (4.5%) 1 (4.5%)
1 (4.5%)
1 (4.5%)
1 (4.5%)
1 (4.5%)
1 (4.5%)
1 (4.5%)
3 (13.6) 1 (4.5%)
1 (4.5%) 1 (4.5%)
1 (4.5%) 1 (4.5%)
1 (4.5%)
Patients who discontinued venetoclax
Progressive disease
Death
Withdrawal by patient Physician decision Other
Duration of Ven treatment, months
Mean (SD) Median (range)
Ven discontinuation at
0 to <6 months*, n (%) 6 to <12 months, n (%) 12 to <18 months, n (%) 18 to <24 months, n (%) ≥24 months
For any reason (n=54)
11.7 (8.0) 11.3 (0.1-24.9)
16 (29.6) 13 (24.1) 9 (16.7) 10 (18.5) 6 (11.1)
Due to AE (n=29)
10.6 (7.1) 11.3 (0.5-23.3)
10 (34.5) 7 (24.1) 6 (20.7) 5 (17.2) 1 (3.4)
Due to PD (n=12)
15.4 (7.8) 16.4 (4.4-24.9)
1 (8.3) 4 (33.3) 1 (8.3) 3 (25.0) 3 (25.0)
*One patient was determined as discontinued during post-combination/follow-up, after dis- continuing Ven 33 days after rituximab discontinuation (which occurred on cycle 2 day 1); total Ven treatment duration was 98 days (63 days after rituximab was dosed). AE,: adverse event; PD: progressive disease; SD: standard deviation;Ven: venetoclax.
haematologica | 2022; 107(1)
137