Incidence and clinical impact of absolute lymphocyte count–only progression in chronic lymphocytic leukemia Authors: Xiancheng Wu, Mazyar Shadman, Jenna Voutsinas, Daniel S. Hippe, Timothy W. Randolph, Chaitra S. Ujjani, Christine Crider, Sara Keshmiri, Stephen D. Smith, Ryan C. Lynch, Mengyang Di, Jennifer J. Huang, Paul S. Martin, Brian G. Till, Ajay K. Gopal Published: 2026 ABSTRACT Key Points•Cumulative incidence of ALC-only progression in CLL is higher with chemoimmunotherapy than with BTKi and BCL2i.•OS and TTNT-D are superior for ALC-only progression compared with other modalities of progression. Introduction Current treatments for chronic lymphocytic leukemia (CLL) are very effective, achieving many years of disease control. However, after first-line therapy, patients with CLL require continued laboratory surveillance, and most will eventually require treatment for relapsed disease. The 2018 International Workshop on CLL (iwCLL) guidelines include an increase in the absolute lymphocyte count (ALC) by ≥50% as a criterion for progressive disease (PD).1 As a result, ALC increase is an important part of the primary end point of progression-free survival (PFS) in many phase 3 trials, equally weighted with other modalities of PD (other PD) such as progressive lymphadenopathy or cytopenias.1, 2, 3 However, ALC-only PD (ALC PD) among patients who have received time-limited chemoimmunotherapy (CIT) have better overall survival (OS) and time to next treatment (TTNT) than other types of PD.4 In clinical practice, ALC PD is often not an indication for treatment and does not warrant initiation of alternative treatment in the absence of other modalities of symptomatic progression. Importantly, multiple large phase 3 trials comparing novel agents to time-limited CIT have shown a significant PFS improvement, but there are no data addressing whether frequencies and outcomes of ALC PD vs non-ALC PD differ in the modern era of targeted therapies.3,5, 6, 7 Furthermore, early progression of disease has been shown to independently predict poor survival in patients with CLL treated with CIT and other lymphoma subtypes.8 More recently, studies have evaluated the role of early PD in patients treated with novel therapy and the most optimal time point for classification of early PD.9,10 However, no studies to date have evaluated the combination of mode of progression and time of progression as a predictor of survival. To address these questions, we retrospectively examined patients with CLL/small lymphocytic lymphoma (SLL) who were treated at the Fred Hutchinson Cancer Center and compared the incidences of ALC PD between different regimens, as well as the clinical outcomes of these patients compared with those with non-ALC PD. We also evaluated the validity of combining early time to progression with mode of progression as a tool for identifying patients at the highest risk of poor survival. Methods Patient eligibility and study design Consecutive patients with CLL/SLL who were treated between 2016 and 2025 at the Fred Hutchinson Cancer Center and University of Washington were collected in an institutional registry and included in this study. Information not available in the registry was collected via manual chart review. Patients who received the following frontline therapy were included: (1) continuous Bruton tyrosine kinase inhibitor (BTKi) monotherapy, with or without anti-CD20 monoclonal antibody; (2) bendamustine-rituximab (BR); and (3) fixed-duration B-cell lymphoma 2 inhibitor (BCL2i) therapy, with or without an anti-CD20 monoclonal antibody and/or BTKi. Clinical, molecular, treatment, progression, and outcome data were collected. Molecular data other than immunoglobulin heavy-chain variable region status were only included if collected before initiation of frontline treatment. PD was defined according to the iwCLL 2018 criteria. Patients with transient lymphocyte increase due to BTKi initiation or restart were not counted as PD events as per iwCLL criteria. Patients who had disease progression before second-line therapy after stopping continuous BTKi due to toxicity or preference were also classified as having PD. Patients who started second-line therapy for toxicity or preference without PD were censored at the start of second-line therapy. PD events were classified as ALC PD or other PD based on retrospective assessment of information available in the electronic medical record. PD was classified as ALC PD if the only iwCLL 2018 criterion met was ALC increase (≥50% increase from nadir with at least 5 × 109/L). Quantification of B-lymphocytes was not required for classification of ALC PD. PD was classified as other PD if any other iwCLL 2018 criteria were met. To account for real-world delays in imaging studies, progressive lymphadenopathy within 2 months of ALC increase were considered other PD. Patients who did not experience any progression events per iwCLL criterial were classified as no PD. Outcomes and measures The primary objective was to compare incidences of ALC PD between the 3 different frontline treatment types and assess for differences in OS and TTNT or death (TTNT-D) between patients with ALC PD and those with other PD. Secondary objectives included evaluation of time from ALC PD to other PD, comparison of OS between nonprogressors, ALC PD, and other PD, and comparison of OS between patients who had other PD within 24 months of treatment initiation compared with patients who did not have early progression or progressed by ALC PD. Early progressors were classified as having progressed within 24 months, based on prior reports of the negative impact of PD within 24 months on survival in CLL.8 Statistical analyses The Kaplan-Meier method was used to analyze TTNT-D and OS from the time of progression after first-line treatment, and comparisons between groups were conducted using the log-rank test. Cox proportional hazards regression was used to estimate hazard ratios (HR) with 95% confidence intervals (CI). Median follow-up was estimated using the reverse Kaplan-Meier method. Continuous variables were summarized using median and interquartile range (IQR). ALC counts were reported in units of ×109/L. Incidence of ALC PD from treatment start was estimated using the cumulative incidence function, with death from any cause and development of other PD as competing risks. Patients were censored at the time of last follow-up without evidence of progression. Comparison of incidence between treatment groups was conducted using Gray test. Cause-specific Cox proportional hazards regression analysis was completed to assess the association between treatment group and hazard of ALC PD. Time-dependent Cox proportional hazards regression was used to compare survival from first-line treatment start between patients who developed ALC PD, those who developed other PD, and those who did not progress, with progression status modeled as a time-dependent covariate and adjusting for frontline treatment type. This approach was chosen to account for potential immortal time bias. A time-dependent Cox model was also used for comparison of early, late, and nonprogressors. Statistical software R version 4.4.1 (R Core Team, Vienna, Austria) was used for statistical analyses. This study was approved by the institutional review boards of the Fred Hutchinson Cancer Center (study number 9777). This study was conducted in accordance with the Declaration of Helsinki. Results Patient, disease, and treatment characteristics Patient baseline and treatment characteristics are listed in Table 1, and molecular profiles are listed in Table 2. A total of 339 patients starting initial therapy were identified, including 179 with continuous BTKi therapy, 66 with fixed-duration BR, and 94 with fixed-duration BCL2i therapy. Patients received first-line therapy between 2011 and 2025, with 12 of 339 patients starting therapy before 2013. All patients were seen at the Fred Hutchinson Cancer Center and University of Washington between 2016 and 2025.Table 1.Baseline and treatment characteristicsCharacteristicAll patientsN = 339ALC PDn = 53Other PDn = 83Age, median (IQR), y65 (58-71)63 (58-69)64 (58-69)Sex, n (%) F123 (36)22 (42)32 (39) M216 (64)31 (58)51 (61)BTKi, n (%)179 (52)26 (49)43 (52) I78 (23)16 (30)22 (27) I + anti-CD2020 (6)4 (8)3 (4) A50 (15)5 (9)8 (10) A + anti-CD2010 (3)1 (2)5 (6) Z21 (6)0 (0)5 (6)BR, n (%)66 (19)21 (40)33 (40)BCL2i, n (%)94 (28)6 (11)7 (8) Ven2 (1)0 (0)1 (1) Ven + anti-CD2069 (20)4 (8)5 (6) Ven + BTKi10 (3)2 (4)0 (0) Ven + BTKi + anti-CD2011 (3)0 (0)1 (1) S + BTKi2 (1)0 (0)0 (0)A, acalabrutinib; F, female; I, ibrutinib; M, male; max, maximum; min, minimum; S, sonrotoclax; Ven, venetoclax; Z, Zanubrutinib.Table 2.Molecular characteristics of patients with PDTreatment groupBTKiBRBCL2iCharacteristicALC PDn = 26Other PDn = 43ALC PDn = 21Other PDn = 33ALC PDn = 6Other PDn = 7IGHV mutation, n (%) Mutated7 (27)7 (16)6 (29)0 (0)2 (33)1 (14) Unmutated10 (38)10 (23)4 (19)10 (30)2 (33)5 (71) Unknown9 (35)26 (60)11 (52)23 (70)2 (33)1 (14)TP53 mutation, n (%) Yes0 (0)2 (5)0 (0)1 (3)1 (17)0 (0) No5 (19)8 (19)3 (14)4 (12)3 (50)3 (43) Unknown21 (81)33 (77)18 (86)28 (85)2 (33)4 (57)Deletion 17p, n (%) Yes7 (27)14 (33)0 (0)1 (3)0 (0)2 (29) No17 (65)21 (49)12 (57)23 (70)6 (100)4 (57) Unknown2 (8)8 (19)9 (43)9 (27)0 (0)1 (14)Deletion 11q, n (%) Yes5 (19)5 (12)3 (14)9 (27)1 (17)1 (14) No18 (69)27 (63)9 (43)15 (45)5 (83)3 (43) Unknown3 (12)11 (26)9 (43)9 (27)0 (0)3 (43)Complex karyotype, n (%) Yes5 (19)6 (14)1 (5)1 (3)2 (33)1 (14) No5 (19)10 (23)6 (29)8 (24)4 (67)3 (43) Unknown16 (62)27 (63)14 (67)24 (73)0 (0)3 (43)IGHV, immunoglobulin heavy-chain variable region. Table 1.Baseline and treatment characteristicsCharacteristicAll patientsN = 339ALC PDn = 53Other PDn = 83Age, median (IQR), y65 (58-71)63 (58-69)64 (58-69)Sex, n (%) F123 (36)22 (42)32 (39) M216 (64)31 (58)51 (61)BTKi, n (%)179 (52)26 (49)43 (52) I78 (23)16 (30)22 (27) I + anti-CD2020 (6)4 (8)3 (4) A50 (15)5 (9)8 (10) A + anti-CD2010 (3)1 (2)5 (6) Z21 (6)0 (0)5 (6)BR, n (%)66 (19)21 (40)33 (40)BCL2i, n (%)94 (28)6 (11)7 (8) Ven2 (1)0 (0)1 (1) Ven + anti-CD2069 (20)4 (8)5 (6) Ven + BTKi10 (3)2 (4)0 (0) Ven + BTKi + anti-CD2011 (3)0 (0)1 (1) S + BTKi2 (1)0 (0)0 (0)A, acalabrutinib; F, female; I, ibrutinib; M, male; max, maximum; min, minimum; S, sonrotoclax; Ven, venetoclax; Z, Zanubrutinib. Baseline and treatment characteristics A, acalabrutinib; F, female; I, ibrutinib; M, male; max, maximum; min, minimum; S, sonrotoclax; Ven, venetoclax; Z, Zanubrutinib. Table 2.Molecular characteristics of patients with PDTreatment groupBTKiBRBCL2iCharacteristicALC PDn = 26Other PDn = 43ALC PDn = 21Other PDn = 33ALC PDn = 6Other PDn = 7IGHV mutation, n (%) Mutated7 (27)7 (16)6 (29)0 (0)2 (33)1 (14) Unmutated10 (38)10 (23)4 (19)10 (30)2 (33)5 (71) Unknown9 (35)26 (60)11 (52)23 (70)2 (33)1 (14)TP53 mutation, n (%) Yes0 (0)2 (5)0 (0)1 (3)1 (17)0 (0) No5 (19)8 (19)3 (14)4 (12)3 (50)3 (43) Unknown21 (81)33 (77)18 (86)28 (85)2 (33)4 (57)Deletion 17p, n (%) Yes7 (27)14 (33)0 (0)1 (3)0 (0)2 (29) No17 (65)21 (49)12 (57)23 (70)6 (100)4 (57) Unknown2 (8)8 (19)9 (43)9 (27)0 (0)1 (14)Deletion 11q, n (%) Yes5 (19)5 (12)3 (14)9 (27)1 (17)1 (14) No18 (69)27 (63)9 (43)15 (45)5 (83)3 (43) Unknown3 (12)11 (26)9 (43)9 (27)0 (0)3 (43)Complex karyotype, n (%) Yes5 (19)6 (14)1 (5)1 (3)2 (33)1 (14) No5 (19)10 (23)6 (29)8 (24)4 (67)3 (43) Unknown16 (62)27 (63)14 (67)24 (73)0 (0)3 (43)IGHV, immunoglobulin heavy-chain variable region. Molecular characteristics of patients with PD IGHV, immunoglobulin heavy-chain variable region. Twenty-six patients (15%) progressed by ALC PD and 43 (24%) by other PD in the BTKi cohort, 21 (32%) by ALC PD and 33 (50%) by other PD in the BR cohort, and 6 (6%) by ALC PD and 7 (7%) by other PD in the BCL2i cohort. The median ALC at the time of PD was 7.55 × 109/L (IQR, 6.17 × 109 to 10.54 × 109/L) among patients with ALC PD and 3.95 × 109/L (IQR, 1.50 × 109 to 11.23 × 109/L) among patients with other PD. The median ALC before treatment start was 39.2 × 109/L (IQR, 11.4 × 109/L to 103 × 109/L) for all patients, 79.8 × 109/L (IQR, 36.4 × 109 to 97.5 × 109/L) for patients with ALC PD, and 22.8 × 109/L (IQR, 7.59 × 109 to 74.3 × 109/L) for those with other PD. Twenty-eight of 83 patients (34%) with other PD also had lymphocytosis at the time of PD, including 12 of 43 (28%) in the BTKi cohort, 13 of 33 (39%) in the BR cohort, and 3 of 7 (43%) in the BCL2i cohort. Among patients who progressed, the most common treatments were ibrutinib (n = 38), acalabrutinib (n = 13), and ibrutinib plus rituximab (n = 7) in the BTKi cohort and venetoclax plus obinutuzumab/rituximab (n = 9) in the BCL2i cohort. The median duration of BCL2i treatment was 12 months (IQR, 12-12). Among patients who developed other PD (n = 83), the most common modes of progression were lymphadenopathy (n = 59) and cytopenia (n = 17). Richter transformation to large BCL occurred in 10 patients. Seventeen patients had SLL based on iwCLL and National Comprehensive Cancer Network guidelines. Among patients who had PD, 13 (10%) were on a clinical trial at the time of progression, including 5 (7%) in the BTKi cohort, 6 (11%) in the BR cohort, 2 (15%) in the BCL2i cohort. Twenty-five patients progressed while off BTKi therapy (24 due to adverse events (AE) and 1 due to personal choice) and before the initiation of second-line treatment, and of these 25 patients, 10 (40%) progressed by ALC PD and 15 (60%) by other PD. Forty-four patients progressed while on BTKi therapy, and of these 44 patients, 16 (36%) progressed by ALC PD and 28 (64%) by other PD. Of all patients who experienced ALC PD (n = 53), 11 patients did not initiate a new line of treatment. Among those who did (n = 42), 13 started next treatment for ALC PD, and 29 started next treatment after subsequent development of other PD. For patients with ALC PD who delayed next line of treatment until development of other PD, the median time from ALC PD to other PD was 10.5 months. The median follow-up times were 70 months (95% CI, 63-76) for all patients, 72 months (95% CI, 62-80) for the BTKi cohort, 102 months (95% CI, 94-130) for the BR cohort, and 45 months (95% CI, 36-53 months) for the BCL2i cohort. Incidence and outcomes of ALC PD vs other PD and no PD The 60-month cumulative incidence of ALC PD from treatment start was 23% (95% CI, 14-34) in the BR cohort, 16% (95% CI, 11-23) in the BTKi cohort, and 15% (95% CI, 5.8-29) in the BCL2i cohort (P = .021; Figure 1). Using a cause-specific Cox proportional hazards regression analysis, BR was associated with a higher hazard of ALC PD than BCL2i (HR, 5.19; 95% CI, 2.08-12.9) and BTKi (HR, 2.43; 95% CI, 1.37-4.31; supplemental Table 1). We also performed a sensitivity analysis excluding 17 patients with SLL and showed that the cumulative incidence of ALC PD remain highest with BR (supplemental Figure 1). The 60-month cumulative incidence of ALC PD was 25% (95% CI, 15-37) for BR, 17% (95% CI, 12-24) for BTKi, and 16% (95% CI, 6.2-31) for BCL2i (P = .023). Cause-specific Cox regression analysis was consistent with higher hazard of ALC PD with BR than BCL2i (HR, 4.98; 95% CI, 1.99-12.5) and BTKi (HR, 2.34; 95% CI, 1.31-4.18; supplemental Table 2).Figure 1.Cumulative incidence. Cumulative incidence of ALC PD between treatment groups. Cumulative incidence. Cumulative incidence of ALC PD between treatment groups. The median OS from the time of progression was 116 months (95% CI, 91 months to not reached) for patients with ALC PD vs 71 months (95% CI, 46 months to not reached) for patients with other PD (P = .0012; Figure 2A). The 5-year estimates of OS were 83% (95% CI, 71-97) for ALC PD and 56% (95% CI, 45-70) for other PD. ALC PD was also associated with numerically higher OS across treatment groups, although statistical significance was not observed in the BR and BCL2i subgroups (P = .067 and P = .09, respectively; Figure 2B-D). In Cox regression analysis, other PD was associated with a higher hazard of death than ALC PD (HR, 3.15; 95% CI, 1.52-6.53; supplemental Table 3). Patients with ALC PD still demonstrated improved OS compared with those with other PD when patients with Richter transformation (median OS, 116 months vs 74 months; P = .0088) and SLL (median OS, 116 vs 71 months; P = .0019) were excluded. The 5-year estimates of OS in ALC PD and other PD were 83% (95% CI, 71-97) and 63% (95% CI, 51-77) with exclusion of Richter transformation and 83% (95% CI, 71-97) and 56% (95% CI, 44-71) with exclusion of SLL. Similarly, Cox regression analysis showed a higher hazard of death with other PD than ALC PD when Richter transformation (HR, 2.62; 95% CI, 1.24-5.55) and SLL (HR, 3.06; 95% CI, 1.46-6.43) were excluded (supplemental Table 4).Figure 2.OS. Comparison of OS between ALC PD and other PD in patients treated with all treatments (A), BTKi (B), BR (C), and BCL2i (D). OS. Comparison of OS between ALC PD and other PD in patients treated with all treatments (A), BTKi (B), BR (C), and BCL2i (D). We also compared OS from the time of treatment initiation using a time-dependent Cox model, with progression status as a time-varying covariate and adjusting for frontline treatment (Table 3). We showed that compared with patients with no PD, ALC PD was associated with intermediate survival (HR, 3.62; 95% CI, 1.48-8.85), and other PD was associated with the worst survival (HR, 11.7; 95% CI, 5.99-22.9).Table 3.Time-dependent Cox model of OS, adjusted for frontline treatment typeCharacteristicn∗HR95% CIP valueMode of progression339 No PD203—— ALC PD533.621.48-8.85.005 Other PD8311.75.99-22.9<.001Mode and time of progression339 No PD203—— ALC PD, >24 mo343.241.08-9.76.037 ALC PD, ≤24 mo194.041.29-12.7.017 Other PD, >24 mo4211.04.71-25.5<.001 Other PD, ≤24 mo4112.05.93-24.1<.001∗The sample size of each group is based on the group reached by the end of each patient’s follow-up. Table 3.Time-dependent Cox model of OS, adjusted for frontline treatment typeCharacteristicn∗HR95% CIP valueMode of progression339 No PD203—— ALC PD533.621.48-8.85.005 Other PD8311.75.99-22.9<.001Mode and time of progression339 No PD203—— ALC PD, >24 mo343.241.08-9.76.037 ALC PD, ≤24 mo194.041.29-12.7.017 Other PD, >24 mo4211.04.71-25.5<.001 Other PD, ≤24 mo4112.05.93-24.1<.001∗The sample size of each group is based on the group reached by the end of each patient’s follow-up. Time-dependent Cox model of OS, adjusted for frontline treatment type The sample size of each group is based on the group reached by the end of each patient’s follow-up. We next evaluated TTNT-D from the time of progression. TTNT-D was significantly longer among patients with ALC PD (median, 14 months [95% CI, 11-22]) vs other PD (median, 2.2 months [95% CI, 1.5-4.1]; P < .0001; Figure 3A). TTNT-D remained longer for patients with ALC PD than other PD across treatment modalities (P < .0095 for each; Figure 3B-D). Similarly, Cox regression analysis of TTNT-D demonstrated a significantly shorter TTNT-D for patients with other PD than those with ALC PD (HR, 3.36; 95% CI, 2.27-4.95; supplemental Table 5).Figure 3.TTNT-D. Comparison of TTNT-D between ALC PD and other PD in patients treated with all treatments (A), BTKi (B), BR (C), and BCL2i (D). TTNT-D. Comparison of TTNT-D between ALC PD and other PD in patients treated with all treatments (A), BTKi (B), BR (C), and BCL2i (D). Mode of progression combined with time to progression as a prognosticator of survival Comparison of early progressors (≤24 months) and late progressors showed numerically longer survival without achieving statistical significance in late progressors, with a median OS of 116 months from the time of PD (95% CI, 116 months to not reached) in late progressors and 89 months (95% CI, 65 months to not reached) in early progressors (P = .2; supplemental Figure 2). Sixty-month estimates of OS were 64% (95% CI, 52-79) in early progressors and 68% (95% CI, 56-83) in late progressors. Cox regression analysis demonstrated a higher hazard of death in early progressors without reaching statistical significance (HR, 1.47; 95% CI, 0.81-2.65). A time-dependent Cox proportional hazards model evaluating the association between OS from the time of treatment initiation and the combination of mode and timing of progression demonstrated an HR of 3.24 for ALC PD >24 months (95% CI, 1.08-9.76), 4.04 for ALC PD ≤24 months (95% CI, 1.29-12.7), 11.0 for other PD >24 months (95% CI, 4.71-25.5), and 12.0 for other PD ≤24 months (95% CI, 5.93-24.1; Table 3). Comparisons of differences in OS between ALC PD ≤24 months and ALC PD >24 months and between other PD ≤24 months and other PD >24 months did not reach statistical significance. Discussion In our retrospective analysis, we showed that compared to other modalities of CLL progression, progression by ALC only portended a more favorable OS and TTNT-D. We also demonstrated that the cumulative incidence of ALC PD was different between treatment modalities: higher with CIT than with BTKi and BCL2i. Pivotal phase 3 trials in CLL have compared frontline BTKi3,5, 6, 7 and BCL2i6 against CIT, with PFS as the primary end point, which does not account for the possible differential incidence of ALC PD between treatment modalities or the unequal impact of ALC PD vs other PD on survival. Increased incidence of ALC PD in CIT over novel therapy may exaggerate the magnitude of PFS benefits observed with novel therapies, without corresponding improvements in OS, due to the comparatively benign prognostic impact of ALC PD. This may in part explain the relevant discordance between PFS and OS in these trials. Similar observations have been made in other areas of oncology, such as in prostate cancer, in which the clinical relevance of biochemical PFS has been questioned,11 and biochemical, radiographical, and clinical progression are often reported separately due to varying frequency and significance.12 Pivotal phase 3 trials in prostate cancer have used radiographic PFS as a primary end point while biochemical and clinical PFS were classified as secondary end points.13 Similarly, studies in multiple myeloma have shown that biochemical relapse predicted longer OS and TTNT than clinical progression.14 Outside of oncology, major adverse cardiovascular events (MACE) is a key composite trial end point in cardiology that includes cardiovascular death, nonfatal myocardial infarction, and stroke.12 Studies routinely report the frequency of the individual components of MACE in recognition of the different clinical significance of each individual components of MACE, a practice that has not translated to composite end point reporting in hematology/oncology.12 Our analyses suggest that ALC PD as a component of PFS in CLL varied in frequency based on treatment and clinical significance compared with other components of PFS and may therefore be considered for separate reporting in future studies. Information on the incidence and outcomes of the individual components of CLL progression should be readily available in prospectively collected databases from large, randomized trials and could be easily reported. We performed additional sensitivity analyses excluding patients who developed Richter transformation to large BCL and patients with confirmed SLL. Patients with Richter transformation are well documented to have significantly worse outcomes than other modalities of progression,15 and patients with SLL may have a lower incidence ALC PD. Our analyses showed that after excluding these groups, ALC PD remained higher with CIT, and other PD predicted worse survival than ALC PD. We also examined the value of combining mode of PD with early vs late PD as a prognosticator of survival. Although a prior study in patients treated with CIT showed that patients who progressed within 24 months had inferior outcomes,8 in our smaller study, evidence for this was observed without achieving the cutoff for statistical significance. Similarly, numerical comparisons revealed that patients with non-ALC PD within 24 months may have the worst prognosis and that patients with ALC PD after 24 months may have the best prognosis, but this observation was largely driven by differences in survival between ALC PD and other PD, and comparisons between PD within 24 months and PD after 24 months did not achieve statistical significance. Future studies with a larger patient sample size will be necessary to further investigate this question. Limitations of our study included its monocentric and retrospective design with a heterogeneous patient population treated over an extended time-period, which may potentiate biases including selection biases and confounding variables. The BCL2i cohort had notably heterogeneous treatment regimens, shorter duration of follow-up than the other cohorts, and low number of progression events, which limit the generalizability of findings in this group. Data regarding tumor molecular features were limited due to missing data or absence of testing. Misclassification of ALC-only progression is possible, because cross-sectional imaging was not uniformly performed at the time of progression to assess for asymptomatic lymphadenopathy. However, this approach is consistent with the iwCLL 2018 guidelines and reflects routine real-world practice. Furthermore, identification and quantification of B-lymphocytes were not required for classification as ALC PD in our study, which is consistent with real-world practice and many clinical trials.3 However, strict interpretation of the 2018 iwCLL guidelines and certain clinical trials protocols require confirmation of B-lymphocytes for determining progression. This highlights an inconsistency in the field, which may differentially affect the timing of ALC PD identification across trials. Additionally, protocol-specific modifications in the definition ALC PD across clinical trials, such as whether ALC increase must be assessed relative to the pretreatment baseline or to the posttreatment nadir, may contribute to heterogeneity in how ALC PD is reported between studies. Patients with CLL-related cytopenias did not consistently undergo confirmation with bone marrow biopsies, and identification of PD events also varied across patients due to varying frequencies of follow-up, physical, and laboratory evaluations, which are all consistent with real-world practice. Our pilot findings should be confirmed using larger databases with prospectively collected data from randomized trials, which should be readily available. In summary, our data suggest that ALC PD portends a better prognosis than other modalities of progression in CLL in patients treated with continuous BTKi and time-limited CIT or BCL2i. Incidence of ALC PD differed between treatment groups and was higher for CIT than novel therapy. Future response criteria may consider reporting PFS both with and without ALC PD to better define the clinical relevance of these results. Conflict-of-interest disclosure: M.S. reports consulting, advisory board, steering committee, or data safety monitoring committee fees from AbbVie, Genentech, AstraZeneca, Genmab, Janssen, BeiGene/BeOne Medicines, Bristol Myers Squibb (BMS), MorphoSys/Incyte, Kite Pharma, Eli Lilly, Fate Therapeutics, Nurix, Merck, Pfizer, and Pierre Fabre; research funding from Mustang Bio, Genentech, AbbVie, BeiGene, AstraZeneca, Genmab, MorphoSys/Incyte, and Vincerx Pharma; stock options with Koi Biotherapeutics; and employment (spouse) with BMS. D.S.H. reports research funding from GE Healthcare. S.D.S. reports research funding from ADC Therapeutics, AstraZeneca, BeiGene, Corvus Pharmaceuticals, Enterome, Genentech, Ignyta (spouse), Incyte Corporation, Kymera Therapeutics, Loxo Oncology, Pfizer Oncology, and Viracta Therapeutics; and consulting fees from Natera and Lumanity. R.C.L. reports research funding from TG Therapeutics, Incyte, Bayer, Cyteir, Genentech, Pfizer, Rapt, Merck, Janssen, and Allogene; and consulting fees/honoraria from Seagen, AbbVie, Janssen, Merck, ADC Therapeutics, Foresight Diagnostics, Genentech, and Natera. J.J.H. reports honoraria from Eisai Inc; consulting fees from AbbVie; and research funding to the institution from BMS. B.G.T. reports research funding from Mustang Bio and BMS/Celgene. A.K.G. reports research funding from Merck, Takeda, AstraZeneca, Agios, Janssen, BMS, Seagen, Teva, Genmab, BeiGene, Pfizer, Umoja, and LTZ Therapeutics; consultancy fees/honoraria from Incyte, Kite, MorphoSys/Incyte, ADC Therapeutics, Acrotech, Merck, BeiGene, Compliment Corporation, Genentech, Lilly, and SciTech; and stock options with Compliment Corporation and SciTech. The remaining authors declare no competing financial interests.