TP53 mutations in CLL: does frequency matter?

In this issue of Blood, Malcikova et al demonstrate that low-burden TP53 mutations with variant allele frequencies (VAFs) between 0.1% and 10% have a significant adverse impact on overall survival (OS), although they do not affect event-free survival (EFS) after first-line chemoimmunotherapy.¹ The authors observed clonal expansion of TP53 mutations, particularly in patients treated with fludarabine, cyclophosphamide, and rituximab, whereas patients managed with watch-and-wait approaches or targeted agents did not show significant clonal expansion.

Aberrations of the TP53 gene, either as a mutation or as deletion 17p, are the most important adverse prognostic markers in chronic lymphocytic leukemia (CLL).² Chromosomal deletions can be detected via fluorescence in situ hybridization, and 10% to 20% is the generally used threshold for del(17p) positive status.³ All established guidelines recommend screening for gene mutations, because 20% of patients without del(17p) can still carry TP53 mutations.⁴ The European Research Infrastructure Consortium (ERIC) guidelines suggest multiple possible methods for detecting TP53 mutations, including Sanger sequencing, denaturing high performance liquid chromatography, and arrays.⁵ In the latest edition of ERIC, next-generation sequencing (NGS) was proposed as the gold standard, given its high sensitivity and ability to detect VAFs down to <1%. However, the clinical impact of VAFs <10% has not been conclusively established; therefore, the most commonly used threshold for reporting variants is set at >10% VAFs.

In the Malcikova study, the authors first looked at patients with active disease at the time of treatment initiation (ie, not at diagnosis, which is when most data are collected^6,7). They observed that patients with TP53 mutations of <1% VAF had a significantly shorter OS than patients with wild-type TP53, whereas patients with VAFs of 1% to 10% had outcomes similar to those in patients without TP53 aberrations. This paradoxical observation highlights one of the pitfalls when comparing data from studies on this topic: the majority of patients in the Malcikova et al analysis received chemoimmunotherapy as a first-line treatment, and not all patients switched to targeted agents at relapse. The authors demonstrate that OS was highly dependent on the use of targeted agents, and therefore a lack of access to these compounds might have affected OS outcomes, irrespective of VAFs. In addition, the independent prognostic value of IGHV gene mutational status is also likely to influence the study results. Hence, the authors systematically worked out the impact of TP53 by adjusting for unmutated IGHV status as well as targeted treatment and showed that low-burden TP53 aberrations of any frequency were associated with shorter OS. Interestingly, the authors were able to demonstrate that low-burden TP53 aberrations expand by almost 15-fold upon treatment with chemoimmunotherapy, whereas almost no clonal expansion was observed upon targeted treatment. This explains why no differences in EFS were detected in the first-line chemoimmunotherapy setting: the adverse clones actually expand and evolve their negative impact after exposure to chemoimmunotherapy, which eventually leads to shorter OS.

The article by Malcikova et al has several clinical implications. First, it clearly demonstrates how targeted treatments are able to improve the prognosis of patients with TP53 mutations of any frequency, suggesting that appropriate genetic screening followed by access to targeted treatment agents in patients with TP53 of any frequency is paramount. It also shows that genotoxic chemotherapy can facilitate expansion of adverse clones and that this process can be limited to a certain extent with targeted agents. Notably, the study highlights that even in the context of targeted treatment, aberrations of the TP53 gene are still relevant adverse prognostic features.

Some limitations of this study need to be considered. First, the retrospective nature of the analysis could be potentially affected by selection bias, in particular because this is a monocentric patient cohort. Moreover, the effect of targeted treatment was explored only in the relapsed/refractory disease setting, but not in patients who received targeted agents as first-line therapy. Furthermore, because the established treatment schemes with targeted agents (eg, continuous single-agent treatment vs combination treatment of a fixed duration) vary substantially, no conclusions can be made on different clonal expansion with different agents (eg, Bruton tyrosine kinase inhibitors vs BCL2 inhibitors).

How should we classify TP53 mutations in the future? The thorough work by Malcikova et al provides strong arguments for lowering the VAF reporting threshold to below 10% to better guide clinical decision making. Data from prospective clinical studies such as CLL13, CLL14, or CLL17, which all use chemoimmunotherapy as well as targeted treatment options and make use of NGS-based TP53 diagnostics, will soon provide further insights into the prognostic value of low-burden TP53 mutations.^8,9 In settings with guaranteed methodologic quality, reporting of VAFs <10% for TP53 should be considered for routine clinical care.