Real-world time to discontinuation of first-line venetoclax + obinutuzumab in chronic lymphocytic leukemia/small lymphocytic lymphoma

Abstract

Objective

To evaluate the time to discontinuation (TTD) and baseline characteristics among patients with chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL) treated with first-line (1L) venetoclax + obinutuzumab (VO) in the United States.

Methods

A nationwide electronic health record-derived database was used to select adults with CLL/SLL initiating a 1L venetoclax-based regimen between April 11, 2016–July 31, 2020. Study measures included TTD (defined as >120-day treatment gap or switching therapy) and baseline characteristics by discontinuation status.

Results

A total of 113 patients receiving 1L VO on/before July 31, 2020 were eligible for analysis (mean age: 65.9 years; 31.9% women). During the first 60 days post-treatment initiation, 3.5% had tumor lysis syndrome (TLS). The proportion of patients using corticosteroids, anti-hyperuricemics, and anti-emetics was higher during the first 60 days post-treatment initiation (100.0%, 78.8%, and 52.2%, respectively) than the period from day 61 onward (67.0%, 45.5%, and 33.9%, respectively). Mean (median) duration of active treatment was 11.6 (12.1) months; 16.8% discontinued treatment before completing 12 cycles, 68.1% completed ≥12 cycles (among which 29.9% completed ≥15 cycles), and 15.0% who did not discontinue treatment were censored before completing 12 cycles. Kaplan-Meier analysis showed that median TTD was 13.8 months. Relative to those completing ≥12 cycles, patients discontinuing treatment before completing the prescribed 12 cycles were older (70.4 vs. 65.1 years) and had poorer renal function (36.8% vs. 13.0% with creatinine clearance <60 mL/min).

Conclusion

A small proportion of CLL/SLL patients who were older and had poorer baseline renal function discontinued 1L VO prior to completing 12 treatment cycles. Additionally, treatment utilization, including medications related to TLS mitigation and management, was more intense during the initiation phase of VO. Further research with longer follow-up to assess long-term outcomes of VO treatment after early discontinuation is warranted.

Keywords:

• Venetoclax
• obinituzumab
• chronic lymphocytic leukemia
• tumor lysis syndrome
• time to discontinuation
• retrospective study

Introduction

The treatment landscape for chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL) has changed dramatically in recent years due to the advent of novel targeted therapies, including inhibitors of B-cell leukemia/lymphoma 2 (BCL-2) and Bruton’s tyrosine kinase (BTK)^1,2. These therapies can be used as monotherapy or in combination with other treatments, such as anti-CD20 monoclonal antibodies (e.g. rituximab, ofatumumab, or obinutuzumab)².

On May 15, 2019, the combination of the BCL-2 inhibitor venetoclax plus obinutuzumab (VO) was approved by the United States (US) Food and Drug Administration (FDA) for previously untreated CLL/SLL³. The approval was based on the results of the CLL14 trial, which showed that a 12-cycle, fixed-duration VO treatment in the first-line (1L) setting significantly prolonged progression-free survival compared with chemoimmunotherapy⁴. Follow-up data also showed that VO led to higher rates of undetectable minimal residual disease⁵.

VO is now a standard 1L treatment option for patients with CLL/SLL, alongside BTK inhibitors ibrutinib and acalabrutinib². However, venetoclax-based regimens have been associated with an increased risk of tumor lysis syndrome (TLS), especially among patients with a high disease burden². TLS is potentially life-threatening and imposes a significant treatment burden on patients^6,7. A previous study in patients with CLL/SLL showed that TLS was associated with substantially higher utilization of healthcare resources in the US, mostly due to inpatient admissions⁷. The risk of venetoclax-associated TLS has resulted in the application of several defined mitigation measures, such as TLS risk assessments and TLS prophylaxis for patients initiating venetoclax-based regimens⁸. Recent clinical trials have also shown that lead-in treatment with ibrutinib prior to the initiation of venetoclax reduced TLS risk^9–11.

Little is known about the real-world outcomes of the 12-cycle, fixed-duration 1L VO treatment. In addition, there is a paucity of information on baseline factors that may affect patients’ ability to complete this treatment. To address this research gap, the current study evaluated the time to discontinuation (TTD) as well as the demographic and clinical characteristics associated with TTD among patients with CLL/SLL treated with 1L VO in the US using a nationwide electronic health record-derived database.

Methods

Data source

This study used the nationwide Flatiron Health electronic health record-derived de-identified database to select patients who initiated 1L venetoclax at community (∼90% of patients) and academic (∼10% of patients) medical centers between April 11, 2016 (i.e. the approval date for venetoclax for CLL/SLL with 17p deletion) and July 31, 2020 (to allow the opportunity to observe ≥12 treatment cycles prior to data cut-off). The Flatiron Health database includes de-identified data from approximately 280 US cancer clinics (∼800 sites of care)^12,13. It is a longitudinal database comprising patient-level structured electronic health records and unstructured data from physicians’ notes and other documents, including demographics, diagnoses, visits, laboratory tests and vitals, medication administration, medication prescriptions and orders, performance status, and insurance data. All data is curated via technology-enabled abstraction. The database included a sample of patients with CLL/SLL meeting the following criteria: CLL/SLL diagnosis confirmed at any time, ≥2 visits at a clinic contributing data to Flatiron Health on or after January 1, 2011, and ≥1 order of an antineoplastic agent on or after January 1, 2011^12,13. Other oncology-specific data elements included lines of therapy, date of diagnosis, disease subtype (e.g. CLL or SLL), cytogenetics/molecular testing results (e.g. 17p deletions, IGHV mutations, CD20 biomarker), disease staging, and Eastern Cooperative Oncology Group performance status. The data were de-identified and subject to obligations to prevent re-identification and protect patient confidentiality. Institutional Review Board approval of the Flatiron Health’s protocol was obtained prior to study conduct and included a waiver of informed consent.

Study design

A retrospective study design was used (Figure 1). The index date was defined as the date of initiation of a venetoclax-based regimen in 1L on or after April 11, 2016. The baseline period was the up-to-12 months period of continuous clinical activity pre-index. The follow-up period spanned from the index date until the earliest of initiation of second-line therapy, end of continuous clinical activity, end of data availability, or death (i.e. no minimum follow-up time was required beyond the first 28 days post-index, which were used to confirm all agents used in combination therapy). For each patient, clinical activity was defined as the range of dates with visits, use of therapies, laboratory tests, vital assessments, Eastern Cooperative Oncology Group assessments, or comorbidity diagnoses (i.e. all dates of service included in the database). The start of clinical activity was the date of the first event observed in the data; the end of clinical activity was the date of the last event observed in the data.

Figure 1. Study design scheme. Abbreviations: 1L, first line; 2L; second line; CLL, chronic lymphocytic leukemia; SLL, small lymphocytic lymphoma; VO, venetoclax + obinutuzumab.

Study population

Patients were included in the study if they had a confirmed CLL/SLL diagnosis documented in the Flatiron Health database; initiated 1L therapy with venetoclax on or after April 11, 2016 and following CLL/SLL diagnosis confirmation; had ≥28 days of clinical activity post-index date (to ensure the index regimen was identified adequately); and were ≥18 years of age as of the index date (Figure 2). Patients were excluded if they had incomplete line of therapy information in the database; ≥1 diagnosis for end-stage renal disease prior to the initiation of 1L therapy; ≥2 diagnoses for other blood cancers ≥30 days apart, evaluated starting 24 months prior to the initiation of 1L therapy until 6 months prior to confirmed CLL/SLL diagnosis; received a clinical study drug during 1L therapy; ≥1 diagnosis indicating participation in a clinical trial pre-index date; or a 1L regimen containing ibrutinib + venetoclax or acalabrutinib + venetoclax (Figure 2). The final study population (N = 113) analyzed included patients initiating 1L VO between April 11, 2016 and July 31, 2020 to allow for the potential observation of ≥12 treatment cycles (i.e. 336 days, corresponding to 12 cycles of 28 days) before the end of data availability.

Figure 2. Identification of CLL/SLL patients treated with VO in 1L. Abbreviations: 1L, first-line; CLL, chronic lymphocytic leukemia; ICD-9/10-CM, International Classification of Diseases-9/10-Clinical Modification; SLL, small lymphocytic lymphoma; VO, venetoclax + obinutuzumab. (1) Confirmation of CLL/SLL diagnosis is available in the Flatiron Health database. Flatiron Health confirms CLL/SLL diagnoses via physician documentation and review of other unstructured documents. (2) For each patient, clinical activity is defined as the range of dates with visits, use of therapies, laboratory tests, vital assessments, Eastern Cooperative Oncology Group assessments, or comorbidity diagnoses (i.e. all dates of service included in the database). The start of clinical activity was the date of the first event observed in the data, and the end of clinical activity is the date of the last event observed in the data. (3) A patient could meet more than one of the exclusion criteria listed and appear under more than one category. (4) Based on ICD-9-CM code 585.6 and ICD-10-CM code N18.6. (5) Based on ICD-9-CM codes: 200–201, 202.0–202.7, 203, 204.0, 204.2, 205–206, 207.0–207.2, 208.0, 208.2; ICD-10-CM codes: C81–C82, C83.1–C83.7, C84, C85.2, C86, C88, C90, C91.0, C91.2–C91.6, C91.A, C92–C93, C94.0–C94.4, C95.0, and C96. (6) Based on ICD-9 CM code: V70.7 and ICD-10 CM code: Z00.6.

Study measures

Treatment utilization patterns and incidence of TLS (identified using International Classification of Diseases [ICD]-9-Clinical Modification [CM]/ICD-10-CM Codes 277.88/E88.3) were evaluated during two phases. The initiation phase was defined as the 60-day period following the initiation of 1L therapy, to account for the 5-week venetoclax ramp-up period starting on day 22 following the index date¹⁴ and potential delays between when the patient actually started treatment and when it was recorded in the data¹⁵. The post-initiation phase was defined as the remainder of the duration of 1L therapy, from day 61 until the end of 1L therapy. Characteristics during the post-initiation phase were evaluated only among patients who completed the initiation phase (i.e. patients with a line of therapy duration > 60 days).

TTD was defined as the time from the index date to the earliest of stopping treatment because of a gap in treatment >120 days (per Flatiron Health’s algorithm to define lines of therapy) or the day before switching to a next line of therapy. Patients not discontinuing treatment were censored at the end of their last available follow-up (i.e. end of continuous clinical activity in the database, end of data availability, or death).

Patient characteristics were described during the baseline period overall and stratified based on VO discontinuation status, i.e. patients who completed ≥12 cycles of VO treatment, patients discontinuing VO treatment before completing 12 cycles, and patients censored before completing 12 cycles of VO treatment. As part of these characteristics, the comorbidity burden was measured based on the Quan-Charlson Comorbidity Index score¹⁶. Renal function variables that have been reported as risk factors associated with TLS⁶ and evaluable in Flatiron Health data, i.e. chronic kidney disease, renal failure, and low creatinine clearance (<60 mL/min), were also measured.

Statistical analysis

Characteristics reported during the baseline and follow-up periods were described using means, standard deviations (SD), and medians for continuous variables, and frequencies and proportions for categorical variables. Kaplan-Meier (KM) survival curves (i.e. time to event analyses) were used to describe median TTD and KM rates of TTD at 3, 6, 9, 12, and 15 months. All analyses were performed using SAS Enterprise Guide 7.1 (SAS Institute, Cary, NC).

Results

A total of 113 patients receiving VO initiated treatment on or before July 31, 2020 and were eligible for the analysis (Figure 2). Among these 113 patients, mean ± SD age was 65.9 ± 10.2 years and 36 (31.9%) were women (Table 1).

Table 1. Baseline characteristics.

	1L VO
	N = 113
Duration of the baseline period (in months), mean ± SD [median]	8.7 ± 4.6 [12.0]
Age, mean ± SD [median]	65.9 ± 10.2 [66.0]
Women, n (%)	36 (31.9)
Year of index date, n (%)
2018	2 (1.8)
2019	44 (38.9)
2020	67 (59.3)
Practice type, n (%)
Community	101 (89.4)
Academic	12 (10.6)
Quan-CCI, mean ± SD [median]	2.4 ± 1.0 [2.0]
Time from CLL/SLL diagnosis confirmation to the initiation of 1L therapy (in months), mean ± SD [median]	38.9 ± 48.1 [20.2]
Baseline medication use, n (%)
Corticosteroids	17 (15.0)
Anti-hyperuricemics	72 (63.7)
Anti-emetics	47 (41.6)
Anti-infectives	28 (24.8)
Risk factors associated with TLS, n (%)
Chronic kidney disease	3 (2.7)
Renal failure	2 (1.8)
Creatinine clearance (mL/min), mean ± SD [median]	86.9 ± 30.3 [82.7]
Creatine clearance <60 mL/min	20 (17.7)
Clinical characteristics, n (%)
Rai stage at initial CLL/SLL diagnosis
0	34 (30.1)
I	20 (17.7)
II	5 (4.4)
III	5 (4.4)
IV	15 (13.3)
Not documented	34 (30.1)
Most recent ECOG performance status
0	53 (46.9)
1	29 (25.7)
2	4 (3.5)
3	2 (1.8)
4	0 (0.0)
Unknown	25 (22.1)

Abbreviations: 1L, first-line; CLL, chronic lymphocytic leukemia; ECOG, Eastern Cooperative Oncology Group; Quan-CCI, Quan-Charlson Comorbidity Index; SD, standard deviation; SLL, small lymphocytic lymphoma; TLS, tumor lysis syndrome; VO, venetoclax + obinutuzumab.

Treatment utilization patterns and incidence of TLS

Among the 113 patients eligible for the analysis, 112 had sufficient follow-up to enter the post-initiation phase. The mean (median) duration of the post-initiation phase was 13.7 (13.7) months. The proportion of patients using corticosteroids, anti-hyperuricemics, and anti-emetics was higher during the initiation phase (113 [100.0%], 89 [78.8%], and 59 [52.2%], respectively) than the post-initiation phase (75 [67.0%], 51 [45.5%], and 38 [33.9%], respectively). During the initiation phase, four (3.5%) patients had a TLS diagnosis; no patients had a TLS diagnosis during the post-initiation phase (Table 2).

Table 2. Treatment utilization patterns and incidence of TLS during the VO initiation and post-initiation phases.

	1L VO	1L VO
	initiation phase	post-initiation phase
	N = 113	N = 112
Duration of phase (in months), mean ± SD [median]	2.0 ± 0.0 [2.0]	13.7 ± 5.5 [13.7]
Number of prescriptions/orders/administrations for venetoclax per month, mean ± SD [median]	4.4 ± 5.5 [2.0]	1.3 ± 2.5 [0.8]
Use of other medications, n (%)
Corticosteroids	113 (100.0)	75 (67.0)
Anti-hyperuricemics	89 (78.8)	51 (45.5)
Anti-emetics	59 (52.2)	38 (33.9)
Anti-infectives	44 (38.9)	57 (50.9)
Patients with a TLS diagnosis	4 (3.5)	0 (0.0)

Abbreviations: 1L, first line; SD, standard deviation; TLS, tumor lysis syndrome; VO, venetoclax + obinutuzumab.

Time to discontinuation

The mean (median) follow-up time for the 113 patients was 16.3 (16.2) months, and the mean (median) duration of active treatment was 11.6 (12.1) months. Nineteen (16.8%) patients discontinued VO treatment before completing 12 cycles; 77 (68.1%) completed ≥12 cycles (among which 23 [29.9%] completed ≥15 cycles); and 17 patients (15.0%) were censored before completion of 12 cycles. Two (1.8%) patients had a next line of therapy during the follow-up period (Table 3).

Table 3. Treatment discontinuation characteristics for patients treated with 1L VO.

	1L VO
	N = 113
Length of follow-up (in months), mean ± SD [median]	16.3 ± 5.0 [16.2]
Duration of active treatment (in months), mean ± SD [median]	11.6 ± 3.6 [12.1]
Duration < 12 cycles^a^,^b, n (%)	36 (31.9)
Duration ≥ 12 cycles^a, n (%)	77 (68.1)
Duration ≥ 15 cycles^c, n (%)	23 (29.9)
Patients with a next line of therapy, n (%)	2 (1.8)
Patients who discontinued treatment before completing 12 cycles with VO^a, n (%)	19 (16.8)
Patients who stopped treatment for >120 days	17 (15.0)
Patients who switched to another therapy (without stopping treatment for >120 days)	2 (1.8)

Abbreviations: 1L, first line; SD, standard deviation; VO, venetoclax + obinutuzumab.

^a The 12 cycles consist of 28 days each to reflect a full course of fixed-duration treatment with VO.

^b Including 17 patients who were censored before completing 12 treatment cycles.

^c The proportion of patients who completed ≥15 cycles was assessed among those who completed ≥12 cycles.

KM analysis of TTD showed that 53 (46.9%) patients either stopped VO treatment for >120 days or switched to another therapy at some point during their follow-up period (either before, on, or after the expected completion date); most patients stopped or switched treatment around the 12-cycle mark as scheduled based on venetoclax prescribing information¹⁴. Median TTD was 13.8 months (Figure 3).

Figure 3. Time to discontinuation of 1L VO therapy. Abbreviations: 1L, first-line; CI, confidence interval; KM, Kaplan-Meier; VO, venetoclax + obinutuzumab. (1) Refers to the population at risk of having the event at that point in time (i.e. patients who have not had the event and have not been lost to follow-up.

Baseline characteristics stratified by discontinuation status

Patients who discontinued 1L VO treatment before completing the prescribed 12 cycles (N = 19) were older relative to patients who completed ≥12 cycles of treatment (N = 77) (mean ± SD age: 70.4 ± 9.4 and 65.1 ± 10.4 years, respectively). The number of patients with baseline use of corticosteroids was four (21.1%) among those who discontinued before completing the prescribed 12 cycles and 12 (15.6%) among those who completed the full 12-cycle treatment. Additionally, a higher proportion of patients with poorer renal function (based on chronic kidney disease and creatinine clearance <60 mL/min) were observed among those who discontinued before completing 12 cycles of treatment, relative to those who completed ≥12 cycles of treatment. Patients censored before completing 12 cycles of VO treatment initiated treatment later in the study period than the other two patient groups and did not complete treatment by the end of data availability; their median time between CLL/SLL diagnosis and initiation of VO was also shorter (3.1 vs. >20 months; Table 4).

Table 4. Baseline characteristics stratified by discontinuation status for patients initiated on 1L VO.

	Patients who completed ≥12 treatment cycles^a	Patients discontinuing treatment before completing 12 treatment cycles^a	Patients censored before completing 12 treatment cycles^a
	N = 77	N = 19	N = 17
Duration of the baseline period (in months), mean ± SD [median]	8.8 ± 4.4 [12.0]	9.6 ± 4.4 [12.0]	6.8 ± 5.1 [8.0]
Age, mean ± SD [median]	65.1 ± 10.4 [66.0]	70.4 ± 9.4 [72.0]	64.4 ± 9.5 [67.0]
Women, n (%)	25 (32.5)	6 (31.6)	5 (29.4)
Year of index date, n (%)
2018	2 (2.6)	0 (0.0)	0 (0.0)
2019	33 (42.9)	11 (57.9)	0 (0.0)
2020	42 (54.5)	8 (42.1)	17 (100.0)
Practice type, n (%)
Community	68 (88.3)	17 (89.5)	16 (94.1)
Academic	9 (11.7)	2 (10.5)	1 (5.9)
Quan-CCI, mean ± SD [median]	2.4 ± 1.0 [2.0]	2.4 ± 0.8 [2.0]	2.4 ± 1.2 [2.0]
Time from CLL/SLL diagnosis confirmation to the initiation of 1L therapy (in months), mean ± SD [median]	42.2 ± 51.5 [21.7]	35.1 ± 38.4 [23.5]	27.9 ± 41.4 [3.1]
Baseline medication use, n (%)
Corticosteroids	12 (15.6)	4 (21.1)	1 (5.9)
Anti-hyperuricemics	50 (64.9)	13 (68.4)	9 (52.9)
Anti-emetics	34 (44.2)	9 (47.4)	4 (23.5)
Anti-infectives	23 (29.9)	1 (5.3)	4 (23.5)
Risk factors associated with TLS, n (%)
Chronic kidney disease	1 (1.3)	2 (10.5)	0 (0.0)
Renal failure	2 (2.6)	0 (0.0)	0 (0.0)
Creatinine clearance (mL/min), mean ± SD [median]	89.9 ± 29.9 [84.8]	71.4 ± 23.9 [70.2]	90.6 ± 34.3 [82.7]
Creatine clearance <60 mL/min	10 (13.0)	7 (36.8)	3 (17.6)
Clinical characteristics, n (%)
Rai stage at initial CLL/SLL diagnosis
0	26 (33.8)	5 (26.3)	3 (17.6)
I	13 (16.9)	3 (15.8)	4 (23.5)
II	2 (2.6)	1 (5.3)	2 (11.8)
III	1 (1.3)	2 (10.5)	2 (11.8)
IV	12 (15.6)	2 (10.5)	1 (5.9)
Not documented	23 (29.9)	6 (31.6)	5 (29.4)
Most recent ECOG performance status
0	37 (48.1)	9 (47.4)	7 (41.2)
1	20 (26.0)	2 (10.5)	7 (41.2)
2	1 (1.3)	2 (10.5)	1 (5.9)
3	2 (2.6)	0 (0.0)	0 (0.0)
4	0 (0.0)	0 (0.0)	0 (0.0)
Unknown	17 (22.1)	6 (31.6)	2 (11.8)

Abbreviations: 1L, first-line; CLL, chronic lymphocytic leukemia; ECOG, Eastern Cooperative Oncology Group; Quan-CCI, Quan-Charlson Comorbidity Index; SD, standard deviation; SLL, small lymphocytic lymphoma; TLS, tumor lysis syndrome; VO, venetoclax + obinutuzumab.

^a The 12 cycles consist of 28 days each to reflect a full course of fixed-duration treatment with VO.

Discussion

The current retrospective electronic health record-based study found that 16.8% of patients discontinued 1L VO before completing the full course (i.e. 12 cycles) of treatment. Patients discontinuing treatment before completing the prescribed 12 cycles were older, were more likely to use corticosteroids at baseline, and had poorer renal function. Furthermore, utilization of anti-hyperuricemics and anti-emetics was higher during the initiation than post-initiation phase, suggesting closer monitoring during treatment initiation, potentially for TLS mitigation and management^6,17, as venetoclax can induce rapid CLL cell death¹⁴, with alterations in blood chemistries occurring within 6–24 h following treatment initiation^4,18. While sample size was small, these descriptive findings collectively suggest that advanced age, a high baseline disease burden, and poor renal function may contribute to discontinuation of 1L VO treatment for CLL/SLL, highlighting potential challenges in completing the recommended full course of VO treatment in the subset of patients with these characteristics.

Prior real-world studies on discontinuation of venetoclax-based treatment mostly included patients with relapsed/refractory (R/R) CLL/SLL, who may have higher discontinuation rates than 1L patients due to disease progression. In a US retrospective study of patients (99% R/R CLL) receiving venetoclax-based regimens, 29% of patients discontinued treatment over a median follow-up of 7 months¹⁹. In another US and UK retrospective study of patients (96% R/R CLL) treated with venetoclax, the discontinuation rate was 40% over a median follow-up of 12 months²⁰. Contrary to the above studies that included mostly R/R patients treated with either venetoclax monotherapy or combination therapies, the current study specifically evaluated TTD for 1L VO treatment among patients with CLL/SLL, thereby providing some of the earliest insights on treatment persistence to this combination in the real-world US clinical practice since its approval on May 15, 2019³. When compared to other targeted therapies for CLL/SLL, the proportion of patients discontinuing 1L VO treatment before completing 12 cycles (16.8%) was found to be slightly higher than what has been observed in phase 3 clinical trials for BTK inhibitors (which was around 10%)^21–23. However, comparing discontinuation rates between these therapies remains challenging, as VO is a fixed-duration treatment, while BTK inhibitors are indicated as continuous treatment until disease progression. Furthermore, it should be noted that differences in patient populations, study designs, and follow-up lengths have precluded the direct comparability of discontinuation rates and TTD across studies and treatments. Thus, future studies comparing persistence to VO vs. BTK inhibitors using the same methodology and adjusting for differences in patient characteristics between treatment cohorts are warranted.

The current study found that older patients and patients with a poorer baseline renal function, which is a risk factor for TLS⁶, may be more likely to discontinue before completing the full course of 1L VO treatment. These findings are corroborated by the literature on reasons for discontinuation for other CLL/SLL treatments^6,24–26 and the literature showing that venetoclax poses a risk for TLS owing to rapid reduction in tumor¹⁴. Obinutuzumab is also associated with TLS risk, in addition to a higher risk of infection and infusion-related reactions, which may affect patients’ ability to stay on treatment^27,28. The risk of TLS is particularly high among patients with advanced age, impaired renal function, and high disease burden^2,14,18. These characteristics are in line with those harbored by patients who discontinued 1L VO prematurely in the current study, suggesting that a high risk of developing TLS may be a factor contributing to treatment discontinuation, in addition to other factors such as disease progression.

Strategies to mitigate the risk of venetoclax-associated TLS have been developed, including the use of dose ramp-up schedule²⁹, patient stratification by TLS risk, administration of TLS prophylaxis, and close monitoring⁶. Furthermore, as the risk of developing TLS is generally greater at the start of treatment when tumor burden is highest, tumor debulking strategies have also been explored as a means to reduce TLS risk³⁰. Venetoclax is currently approved for CLL/SLL in the 1L setting in combination with obinutuzumab¹⁴, with the latter agent given first for 3 weeks to reduce tumor burden and therefore TLS risk^4,5. Emerging pre-clinical and clinical evidence suggests ibrutinib is also a candidate for this strategy. Through inhibiting BTK, ibrutinib blocks the interaction of CLL cells with the protective lymph node microenvironment and mobilizes the cells away from the lymph node into the peripheral blood stream³¹. As such, treatment with ibrutinib in vivo has been shown to result in notable reduction in lymph node sizes and lymphocyte counts³², reductions which are associated with a reduced need for TLS monitoring/prophylaxis as per the criteria in the venetoclax prescribing information, suggesting lower TLS risk¹⁴. Currently, the combination of ibrutinib and venetoclax is being used as a fixed-duration treatment in previously untreated patients with CLL, with the goal of reducing TLS risk while achieving synergistic efficacy^9–11. In prospective phase 2 and phase 3 trials of 1L ibrutinib and venetoclax among patients with CLL, lead-in single-agent ibrutinib downgraded the risk of TLS prior to the initiation of the ibrutinib plus venetoclax combination treatment^9–11, confirming a prior post-hoc analysis of several CLL clinical trials, in which single-agent ibrutinib resulted in a decline in bulky lymph node disease and absolute lymphocyte count, with an overall downgrade of TLS risk categories among patients³³. Thus, these trial data have supported the use of ibrutinib as lead-in treatment for tumor debulking in CLL. Overall, novel strategies are warranted to mitigate venetoclax-associated TLS risk, which may also help improve treatment persistence to venetoclax-based regimens in a subset of patients with CLL/SLL with a high baseline disease burden or poor renal function, who may be at risk of developing TLS.

Findings of this study should be considered in light of some limitations. First, this study used data derived from electronic health records, which may contain omissions or inaccuracies. Second, the Flatiron Health database has limited information on comorbidities, laboratory results (e.g. absolute lymphocyte count, lymph node size), and healthcare resource utilization; therefore, even if the patient was reported as not having a condition or receiving a healthcare service, it might reflect missing information and not necessarily indicate an absence of the condition or service. Third, given the small sample size, the analyses were descriptive, and no significance testing was performed. Fourth, the length of follow-up was limited by data availability, and few patients discontinued treatment or had a next line of therapy. Fifth, reasons for discontinuation of VO prior to completing 12 cycles of treatment or continuation of VO beyond the prescribed 12 cycles were not available in the data and may be a consideration for future analyses in databases where this information may be available. Finally, the data mostly consisted of patients treated in community practice; hence, findings may not be generalizable to the entire US population of CLL/SLL patients treated with venetoclax.

Conclusion

In this descriptive analysis, a small proportion of patients with CLL/SLL discontinued 1L VO prior to completing 12 cycles of treatment; these patients were older and had poorer baseline renal function (based on chronic kidney disease and creatine clearance <60 mL/min). Additionally, treatment utilization, including medications related to TLS mitigation and management, was more intense during the initiation phase of VO. Further research with longer follow-up to assess long-term outcomes of VO treatment after early discontinuation is warranted.

Transparency

Declaration of funding

This study was funded by Janssen Scientific Affairs, LLC. The sponsor was involved in the study design; data collection, analysis, and interpretation; manuscript writing; and the decision to publish the article.

Declaration of financial/other relationships

XL, ZPQ, and QH are employees of Janssen Scientific Affairs, LLC and stockholders of Johnson & Johnson. LHW was an employee of Janssen Scientific Affairs, LLC at the time the study was conducted. BE, AH, MHL, and PL are employees of Analysis Group, Inc., a consulting company that has provided paid consulting services to Janssen Scientific Affairs, LLC. KAR has received research funding from Genentech, AbbVie, Novartis, and Janssen (not for the present study); consulting fees from Acerta Pharma, AstraZeneca, Innate Pharma, Pharmacyclics, Genentech, and AbbVie; and travel funding from AstraZeneca. SPF and SL were employees of Analysis Group, Inc. at the time the study was conducted.

A reviewer on this manuscript declared that they have received consulting fees with AbbVie, Janssen, BeiGene, and Astrazeneca. Peer reviewers on this manuscript have received an honorarium from CMRO for their review work but have no other relevant financial relationships to disclose.

Author contributions

XL: Conceptualization, investigation, methodology, project administration, resources, supervision, validation, visualization, writing – original draft, writing – review & editing. BE, MHL, and PL: Conceptualization, data curation, formal analysis, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing – original draft, writing – review & editing. ZPQ: Investigation, methodology, project administration, supervision, validation, visualization, writing – review & editing. LHW: Investigation, methodology, validation, visualization, writing – review & editing. SPF, AH, and SL: Data curation, formal analysis, investigation, methodology, software, supervision, validation, visualization, writing – original draft, writing – review & editing. QH: Conceptualization, funding acquisition, investigation, methodology, project administration, resources, supervision, validation, visualization, writing – review & editing. KAR: Conceptualization, investigation, methodology, supervision, validation, visualization, writing – review & editing

Ethical approval

The data were de-identified and subject to obligations to prevent re-identification and protect patient confidentiality. Institutional Review Board approval of the Flatiron’s protocol was obtained prior to study conduct and included a waiver of informed consent.

Previous presentation

Part of the material in this manuscript was presented as a poster at the NCCN 2022 Annual Conference held on March 31–April 2, 2022 in Orlando, FL.

Acknowledgements

Medical writing assistance was provided by Flora Chik, PhD, MWC, an employee of Analysis Group, Inc., a consulting company that has provided paid consulting services to Janssen Scientific Affairs, LLC, which funded the development and conduct of this study and manuscript.

Data availability statement

The data that support the findings of this study have been originated by Flatiron Health, Inc. Requests for data sharing by license or by permission for the specific purpose of replicating results in this manuscript can be submitted to [email protected].