Page 148 - 2022_01-Haematologica-web

A.R. Mato et al.
1.0%]) are presented in Table 4. There were no significant baseline demographic predictors for dose reduction due to venetoclax related AE (Online Supplementary Table S4) and logistic regression modeling was not performed for dose reduction due to AE that were not related to venetoclax, due to small patient numbers (n=3). Median time to first onset of AE leading to venetoclax dose reduction was 3.99 months (range, 0.3–20.5).
Relative dose intensity
Median relative dose intensity (total dose received by patients ÷ expected total target dose, starting from first day of VenR combination therapy until the last day of veneto- clax treatment or clinical cut-off, whichever is earlier) was 97.4% (range, 26.4–100.0; n=189). Relative dose intensity was classified into four categories; distribution of PFS were similar among these categories (Table 7), indicating that there was no compounded effect of both dose reduction and interruption. This analyses was not performed for OS as modeling was insufficiently powered, with only six deaths among the patients who completed venetoclax treatment.
Cumulative exposure
Cumulative treatment (number of 28-day cycles) received by patients (excluding those discontinuing for PD) was statistically significantly associated with reduced risk of both PFS and OS events (PFS: n=167, HR 0.93, 95% CI: 0.88–0.99, P=0.0263; OS: n=161, HR 0.85, 95% CI: 0.78– 0.92, P<0.0001).
Discussion
The results of these analyses highlight the importance of appropriately managing treatment modifications to ensure optimal outcomes for patients receiving targeted treatment for CLL. Treatment discontinuation prior to completing the MURANO regimen (fixed duration of 2 years of vene- toclax) was associated with impaired PFS, while improved PFS and OS were associated with greater cumulative venetoclax treatment exposure. PFS was also statistically significantly lower with discontinuation specifically due to AE. The lack of a significant effect of treatment interruption and dose reduction on outcome highlights the importance of management of AE to allow alternative options such as dose modification or re-initia- tion of therapy following a treatment break, to maximize the duration of therapy.
Early discontinuation of venetoclax was reported in 28% of patients participating in the MURANO study,
similar to the 35% discontinuation rate (median follow- up of 10 months) reported in a real-world study of vene- toclax treatment.32 It is not clearly indicated whether dis- continuation in that study was premature discontinuation (as in the MURANO study) or whether some of these patients were receiving venetoclax monotherapy, where treatment was until progression. The most common rea- son for discontinuation in the real-world study was PD (17%), with 4% reporting discontinuation due to toxicity.32 In contrast, the MURANO data indicate that most patients who discontinue venetoclax prematurely do so due to AE (14.9%), with PD being the second most common cause of discontinuation (6.2%). This empha- sizes the difficulty of assessing the impact of treatment discontinuation due to toxicity or PD based on clinical studies and real-world data, especially given that there are currently insufficient real-world data to assess how venetoclax discontinuation due to toxicity affects out- comes.32 Furthermore, differences in data reported for the MURANO study compared with real-world data reflects the limitations of cross-comparing data without account- ing for differences in patient population and treatment schedule.
In the MURANO study, treatment interruption occurred in 137 of 194 (70.6%) of patients. Venetoclax treatment interruption (for ≥1, ≥8, ≥14 and ≥21 consecu- tive days) showed no statistically significant effect on clinical outcomes (PFS and OS). The rate of interruptions occurring in MURANO is higher than the rate reported in two real-world studies of venetoclax (41% [including patients with dose reductions] and 32%, respectively).33,34 Two explanations potentially account for this difference: firstly, the more stringent reporting of interruptions with- in a clinical trial; and secondly, the stringent protocol requirements for mild asymptomatic neutropenia man- dated for the MURANO study (see the Online Supplementary Appendix), which might contrast with real- world management, where treatment is typically contin- ued with the addition of granulocyte colony-stimulating factor support. Real-world studies of venetoclax provide further evidence that temporary interruption of treatment (length of interruption not disclosed for one study; medi- an 7 days [range, 1–132] for the other) does not have an impact on PFS.33,34
Dose reduction occurred in 45 of 194 (23.2%) of patients in the MURANO study and had no statistically significant effect on PFS or OS. The number of patients requiring dose reductions in MURANO is similar to the 29% reported in the Roeker et al. real-world study of venetoclax.30 Furthermore, the real-world studies of vene- toclax support the findings from MURANO that dose
Table 7. MURANO: landmark analysis of progression-free surviva by venetoclax relative dose intensity quartiles.
Events, n (%)
Kaplan–Meier median, months (95% CI) HR (95% CI)
P-value
Min (26.4%) – <Q1 (93.6%) n=33
8 (24.2) NE (22.9–NE) 1.0
1.0
Q1 (93.6%) – <Median (98.1%) n=35
7 (20.0)
NE (28.1–NE) 0.57 (0.13–2.49) 0.4575
Median (98.1) – <Q3 (99.5%) n=34
9 (26.5) 27.3 (18.8–NE) 1.01 (0.20–5.01) 0.9952
Q3 (99.5%) – Max (100.0%) n=35
11 (31.4) 27.7 (22.3–NE) 0.95 (0.28–3.26) 0.9331
The landmark analysis was performed to study the effect of relative dose intensity on progression-free survival (PFS).The patients who completed venetoclax treatment and had not progressed or were censored at the last dose of venetoclax,were included.The PFS was calculated from the last dose of venetoclax to the first occurrence of progression or death from any cause. CI: confidence interval; HR: hazard ratio; Max: maximum; Min: minimum; NE: not estimable; PFS: progression-free survival; Q: quartile.
140
haematologica | 2022; 107(1)