Historically, chemotherapy had been the most widely used option for the treatment of CLL, which was subsequently replaced by CIT based on the results of practice-changing clinical studies [49]. The most used chemotherapy regimens were initially based on monotherapy with alkylating agents, such as chlorambucil and bendamustine, or purine analogues, namely fludarabine and cladribine [4]. As for combination therapies, the most adopted was the FC regimen (fludarabine and cyclophosphamide), which granted favorable results compared to monotherapies [4]. CIT for CLL consists of the combination of chemotherapy and anti-CD20 mAbs in order to obtain a synergistic effect against tumor cells [13]. The CLL8 phase III randomized trial compared the chemotherapy regimen FC versus FCR in fit CLL patients, demonstrating a significant superiority of the chemoimmunotherapeutic approach [27,50]. Specifically, FCR outperformed FC in both median PFS (56.8 vs. 32.9 months, respectively) and overall survival (OS, not reached vs. 86.0 months, respectively) with a comparable toxicity profile [27].

Different CIT regimens for CLL patients with comorbidities were evaluated in the CLL11 phase III randomized trial, where the Chl-O regimen showed better median PFS and OS compared to chlorambucil plus rituximab or chlorambucil monotherapy [28,51]. The adoption of CIT regimens for CLL has led to the identification of several biomarkers of refractoriness to this therapeutic approach, including the unmutated status of immunoglobulin heavy-chain variable (IGHV) genes and genetic lesions of TP53, BIRC3 and NOTCH1 [24]. In order to overcome resistance to CIT, pathway inhibitors were adopted for the treatment of CLL, with practice-changing results obtained by BTKi and BCL2i [49].

Mature B cells express the B cell receptor (BCR) on their surface, a key component for antigen recognition and B cell activation, composed of an immunoglobulin (Ig) and a signaling subunit [52]. In order to expose an Ig on the external cell membrane, B cells must perform a genetic recombination of the variable Ig genes through a process termed V(D)J rearrangement, which ensures a very high degree of heterogeneity in the BCR repertoire [53]. After antigen recognition, naïve B cells move to lymph node germinal centers (GCs), where somatic hypermutation (SHM) of IGHV genes takes place, potentially increasing the BCR affinity for the recognized antigen [53,54].

Based on the mutational status of IGHV genes, CLL can be divided into two molecular subgroups: (i) IGHV unmutated CLL (U-CLL, ~40% of all CLL), which reflect mature B cells that have not experienced the GC reaction and have undergone maturation in a T-cell-independent manner; and (ii) IGHV mutated CLL (M-CLL, ~60% of all CLL), which reflect mature B cells that have experienced the GC reaction and have undergone the SHM process [25,55,56]. In particular, to be considered as M-CLL, the threshold used in the clinical practice is a deviation in ≥2% of the patient’s IGHV sequence from the germline nucleotide sequence [57]. Unmutated IGHV genes associate more commonly with progressive or R/R CLL, while mutated IGHV genes are more frequently detected in asymptomatic or treatment-naïve disease [58,59]. Importantly, unmutated IGHV genes occur in up to 60% of CLL, requiring treatment according to guidelines.

Beyond its prognostic value, IGHV mutational status is also a predictive biomarker, as shown by the lower response of U-CLL to all the available CIT regimens when compared to M-CLL [27,60,61,62]. Clinical trials evaluating continuous treatment with ibrutinib, acalabrutinib and zanubrutinib have displayed favorable efficacy outcomes in U-CLL, superimposable to those reached with M-CLL, overcoming treatment refractoriness due to IGHV mutational status [63,64,65].

TP53, located on the short arm of chromosome 17 (17p), is an onco-suppressor gene encoding the p53 protein, also called “the guardian of the genome”, which exerts a proapoptotic function in response to DNA damage [66]. Consistently, the disruption of TP53 results in increased resistance to apoptosis induced by DNA-damaging agents, including chemotherapy and, by extension, CIT [67]. Somatic mutations are the most common genetic lesions of TP53 in CLL, followed by del(17p) [68,69].

The disruption of TP53 has been found in 4% to 8% of newly diagnosed CLL, while, as the disease progresses, the frequency of TP53 abnormalities rises, reaching a prevalence of 10–12% at the time of first treatment requirement, ~40% in patients refractory to fludarabine, and 50–60% in those who develop RS [69]. Consequently, genetic lesions of TP53 can be defined as both prognostic and predictive biomarkers. Among patients treated with FCR, the CLL8 trial reported a median PFS of 15.4 months and a median OS of 49 months in TP53-mutated patients, while the median PFS and OS in TP53 wild-type patients were 59 months and not reached, respectively [27]. Similar unfavorable outcomes were reached with the BR and Chl-O regimens in TP53-disrupted CLL, while BTKi-based therapies have obtained remarkable results, which are comparable with those of TP53 wild-type patients [27,28,61,63,64,65].

Due to the significant clinical impact exerted by TP53 disruption, the iwCLL guidelines recommend testing del(17p) via fluorescence in situ hybridization (FISH) and TP53 mutation status via DNA sequencing before every line of treatment [14]. In addition to these recommendations, the European Research Initiative on CLL (ERIC) endorses the possible use of next-generation sequencing (NGS) for TP53 mutation testing since this methodology is characterized by a higher sensitivity compared to traditional Sanger sequencing [70].

The BIRC3 gene has been found to be mutated or deleted in 2–6% of CLL cases [31,32,33]. BIRC3 encodes for the protein c-IAP2, which negatively regulates the MAP3K14 kinase (or NIK–NF-κB-inducing kinase), the key activator of the noncanonical NF-κB signaling pathway, leading to the transcription of genes linked to cell proliferation and survival [71]. Additionally, the aberrant activation of NF-κB signaling in c-IAP2 knockdown models has been shown to increase p53 degradation via the E3 ubiquitin ligase MDM2 [72]. Therefore, the mutational inactivation or deletion of BIRC3 in CLL results in constitutive NF-κB pathway activation, providing pro-survival signals to the leukemic clone, e.g., through the up-regulation of several anti-apoptotic genes, as demonstrated in ex vivo models [34,73].

A retrospective evaluation of the outcome of FCR-treated CLL patients showed a similar median PFS rate between BIRC3- and TP53-disrupted CLL patients (2.2 and 2.6 years, respectively), significantly inferior to the PFS of BIRC3 wild-type patients [34]. Moreover, the CLL14 phase III randomized trial, evaluating front-line treatment with venetoclax plus obinutuzumab versus Chl-O in CLL, has shown poor outcomes in BIRC3 mutated patients treated with Chl-O, with a median PFS of 16.8 months [35,74]. However, ibrutinib- and/or venetoclax-based therapies appear to overcome the resistance conferred by BIRC3 disruption [34,74,75,76,77].

NOTCH1 codes for the transmembrane protein NOTCH1, which acts as a surface receptor for ligands of the SERRATE/JAGGED or DELTA families [78,79]. After being cleaved by γ-secretase, the active subunit of the receptor migrates into the nucleus and acts as a transcription factor for genes involved in cell survival and proliferation, including MYC and components of the NF-κB pathway [37,78]. In CLL, NOTCH1 mutations disrupt the PEST domain, responsible for the promotion of proteasomal degradation of the NOTCH1 protein, resulting in the aberrant activation of the receptor [80,81].

At diagnosis, ~8% of CLL patients harbor a NOTCH1 mutation, but the prevalence of this genetic lesion rises in fludarabine-refractory CLL and RS patients (20.8% and 31.1%, respectively) [38]. Furthermore, genetic lesions of NOTCH1 are thought to be involved in resistance to CLL immunotherapy [79]. The predictive value of NOTCH1 mutations for the treatment with an immunotherapeutic agent has been investigated in the above-mentioned CLL8 trial [27]. In particular, NOTCH1-mutated CLL showed no improvement due to the addition of rituximab since the 5-year PFS rate of NOTCH1-mutated patients was 25.8% for the FC cohort and 26.7% for the FCR cohort (p-value of 0.974) [27]. Similar results were obtained by the COMPLEMENT 1 study, a phase III randomized trial that compared chlorambucil alone versus chlorambucil plus the anti-CD20 mAb ofatumumab, highlighting the role of NOTCH1 mutations in predicting refractoriness to anti-CD20 mAb-based immunotherapy [39,82].

The proposed mechanism for NOTCH1-mediated resistance appears to be linked to the HDAC-mediated repression of the surface exposure of CD20 in NOTCH1-mutated CLL cells, as shown by in vitro models [83]. Remarkably, the CLL11 trial demonstrated a clear superiority of Chl-O over chlorambucil alone, showing better PFS and OS in the Chl-O arm, independent of the presence of NOTCH1 genetic lesions [28]. Hence, these data suggest that the higher clinical efficacy of obinutuzumab may overcome the effect of NOTCH1 mutations. Pathway inhibitors represent a viable option to overcome NOTCH1-mediated refractoriness to CIT and immunotherapy, as shown in the RESONATE phase III randomized trial, where no difference in PFS was detected between NOTCH1 mutated and wild-type CLL treated with ibrutinib [63].