Quest of biomarkers for venetoclax-treated CLL

In this issue of Blood, Roberts et al expand our understanding of prognostic biomarkers predicting the duration of remission in relapsed/refractory chronic lymphocytic leukemia (CLL) treated with venetoclax.¹ 

The BCL2 inhibitor venetoclax, the BTK inhibitor ibrutinib, and the phosphatidylinositol 3-kinase inhibitors idelalisib and duvelisib share similar indication in relapsed/refractory CLL. In the absence of head-to head comparisons, the selection of the most appropriate novel agent to be used in a given patient with relapsed/refractory CLL is currently tailored by general patient consideration, including concomitant medical conditions, preference of continuous vs fixed duration therapy, and drug accessibility. Methodologically sound and independently validated biomarker studies performed in conjunction with clinical trials can help predict which relapsed/refractory CLL subpopulation will benefit more from treatment with venetoclax than with ibrutinib, idelalisib, or duvelisib. Despite evidence from retrospective studies only (ie, low-grade evidence), biomarker studies can provide guidance for the selection of the most appropriate agent for a given patient based on their disease characteristics.

Clinicians should interpret ancillary biomarker studies with caution until the findings are validated in independent trials and/or real-world cohorts. Indeed, biomarker studies conducted in conjunction with clinical trials can be influenced by confounding factors such as significant selection bias. Although several statistical methods can mitigate this bias (eg, multivariate regression analysis), these can only be applied to variables that are known and measurable. The potential for residual confounding, therefore, remains a major limitation and can lead to significant overestimates/underestimates of effect size. In addition, clinical trials, in particular phase 1 and 2 trials, have stringent eligibility criteria that may exclude several important segments of the population (eg, elderly patients, patients with comorbidity, or socioeconomic barriers to care), which may result in limited external validity (ie, generalizability).

The study by Roberts et al is designed as a retrospective observational analysis of the relationship between health-related clinical data collected from 4 pooled phase 1 or 2 venetoclax-based clinical trials and treatment outcome in the setting of relapsed/refractory CLL. The key finding of Roberts et al is that durable response to venetoclax-based therapy is more likely in patients having minimal adenopathy, wild-type TP53 and NOTCH1 genes, and mutated IGHV genes before starting venetoclax, and in patients achieving undetectable minimal residual disease (MRD) on venetoclax.¹  Independent real-world data collected outside of clinical trials have consistently identified TP53 status as a predictor of progression-free survival (PFS) with venetoclax treatment, thus supporting the findings by Roberts et al.²  Conversely, the association between IGHV mutation status and PFS has not been replicated in real-world data, whereas lymph node size, NOTCH1 mutation status, and MRD have not been tested yet as prognostic biomarkers in other cohorts.² 

TP53 status stands as the sole independent and validated prognosticator of progression in independent venetoclax and ibrutinib studies (see table).^1-8  Unfortunately, it is not a useful biomarker for selecting between venetoclax and ibrutinib treatment of relapsed/refractory CLL because having TP53 abnormalities points in the same direction (ie, inferior progression-free survival) with both venetoclax and ibrutinib.^1-8  Although TP53 status does not apparently affect PFS under idelalisib treatment, this result is from a single study and lacks independent validation.⁹ 

Venetoclax, ibrutinib, idelalisib, and duvelisib have been developed in trials as a continuous treatment.^1,8-10  The implications of this continuous approach on long-term safety, drug interactions, quality of life, compliance to therapy, and economic sustainability are not clear. The trend now for treatment of CLL is for a fixed duration or MRD-driven end point using combinations of agents capable of producing MRD-negative responses in a high proportion of cases. Undetectable MRD at the end of induction treatment is a strong indicator of response durability and survival in CLL. MRD status might thus represent a dynamic biomarker to guide treatment duration and to allow patients achieving a negative status to gain drug-free intervals. The study by Roberts et al shows in a powered sample size that undetectable MRD is a surrogate for long-term outcome in the setting of venetoclax. Using treatment with ibrutinib, idelalisib, or duvelisib, undetectable MRD is rare in relapsed/refractory CLL.^8-10  Response depth is irrelevant and has no effect on response duration of BTK and phosphatidylinositol 3-kinase inhibitors unless these agents are discontinued.^8-10  Thus, patients receiving ibrutinib, idelalisib, or duvelisib have to receive treatment until intolerance or disease progression, and cannot have drug-free intervals. The study by Roberts et al shows that undetectable MRD occurs in a sizable proportion of patients. As in the setting of chemoimmunotherapy, achievement of undetectable MRD translates into durable remissions in venetoclax treatment even after stopping therapy, suggesting that venetoclax can be a cornerstone of the next-generation fixed duration or MRD-driven duration therapies.

The study by Roberts et al signals that, in contrast to ibrutinib, bulky lymph nodes is a biomarker of less durable remission with venetoclax treatment. Venetoclax and ibrutinib have therefore complementary activity in clearing the disease across anatomical compartments. Ibrutinib is more active in lymph nodes than blood, where a small lymphocytosis might persist despite continuous treatment. Conversely, venetoclax is more active in blood and marrow where it can deepen the response to undetectable MRD, rather than in lymph nodes. These observations are a strong rationale for testing in venetoclax-ibrutinib combinations clinical trials.