A genetic profiling guideline to support diagnosis and clinical management of lymphomas

Abstract

The new lymphoma classifications (International Consensus Classification of Mature Lymphoid Neoplasms, and 5th World Health Organization Classification of Lymphoid Neoplasms) include genetics as an integral part of lymphoma diagnosis, allowing better lymphoma subclassification, patient risk stratification, and prediction of treatment response. Lymphomas are characterized by very few recurrent and disease-specific mutations, and most entities have a heterogenous genetic landscape with a long tail of recurrently mutated genes. Most of these occur at low frequencies, reflecting the clinical heterogeneity of lymphomas. Multiple studies have identified genetic markers that improve diagnostics and prognostication, and next-generation sequencing is becoming an essential tool in the clinical laboratory. This review provides a “next-generation sequencing” guide for lymphomas. It discusses the genetic alterations of the most frequent mature lymphoma entities with diagnostic, prognostic, and predictive potential and proposes targeted sequencing panels to detect mutations and copy-number alterations for B- and NK/T-cell lymphomas.

Introduction

Mature lymphoid malignancies (Hodgkin (HL) and non-Hodgkin lymphomas (NHL)) are the most common hematological solid neoplasias. Thanks to the implementation of high-throughput molecular analysis, our knowledge of the molecular characteristics of lymphomas has been strengthened. Lymphoma classification is still mainly based on morphology, immunophenotype, and a few genetic characteristics. However, the new classifications of mature lymphoid neoplasms, the International Consensus Classification (ICC) of Mature Lymphoid Neoplasms [1] and the 5th World Health Organization (WHO) Classification of Lymphoid Neoplasms [2], have incorporated newly developed technologies to improve lymphoma classification and genetic alterations are now part of the criteria applied to lymphoma diagnosis.

Standardizing these high-throughput techniques to enable their full and successful clinical application is ongoing, but much controversy persists about the sequencing approach. One of the most critical issues is the method of choice and the composition of the sequencing panel. An ideal panel should be helpful in diagnosis, prognostication, therapy selection, and monitoring but small enough to be broadly and uniformly used. The ability to sequence the whole exome or genome is growing steadily. Nevertheless, a custom panel is currently the most accessible option to broaden the applicability of this approach. These targeted panels enable the analysis of a small number of genes in greater depth, with increased sensitivity and at a lower cost. Amplicon or capture-based sequencing panels could be used. The capture-based ones enable detection not only of single nucleotide variations (SNVs) and indels (insertions and deletions) but also of copy-number alterations (CNAs) and some structural variants. However, the detection of structural alterations needs further improvement. Additionally, standardization of sample management, panel composition, sequencing procedures, bioinformatic analysis, and variant interpretation is essential to produce a useful clinical tool.

This review aims to summarize the fundamental molecular characteristics of the various lymphoma types and to describe the genes relevant to each subtype that could be included in a massive sequencing or next-generation sequencing (NGS) panel.

Mature B-cell lymphomas

Chronic lymphocytic leukemia

Chronic lymphocytic leukemia (CLL) is a low-grade lymphoproliferative disorder characterized by the clonal proliferation and accumulation of mature, typically CD5 + B-cells within the blood, bone marrow, lymph nodes, and spleen [1–4].

More than 80% of CLL cases feature some cytogenetic abnormality and their detection by fluorescence in situ hybridization (FISH) stratify the patients into different risk groups [5]: deletion on the long arm of chromosome 13 (del(13q)), occurs in approximately 55% of cases [6, 7]; trisomy 12 is the second most frequent recurrent chromosomal aberration (10–20% of cases) [6, 8]; deletions on the long arm of chromosome 11 (del(11q)) are found in approximately 25% of chemotherapy-naïve patients with advanced disease and 10% of patients with early disease [9]; and deletion on the short arm of chromosome 17 (del(17p)) occurs in 5–8% of chemotherapy-naïve patients. Only the latter abnormality is considered a significantly negative prognostic factor [10–12]. Other frequent, recurrent abnormalities in CLL include 6q deletion (5%) and 2p gain (5–16%), among others [1, 13].

Somatic hypermutation (SHM) of the immunoglobulin heavy chain variable region (IGHV; < 98% similarity of the IGHV sequence, mutated CLL, M-CLL) confers a better prognosis than the absence of mutation (unmutated CLL, U-CLL) [5, 14, 15]. Recently, the mutation IGLV3-21^R110 found in around 5–15% of CLL, can confer a poor prognosis, independently of the IGHV mutational status [16]. TP53 mutations are found in 4–37% of CLL cases. They may occur alone or, more frequently, in combination with del(17p). They have been associated with chemo-refractoriness and reduced overall survival (OS) [17, 18]. NGS studies have helped identify mutations in other genes with prognostic relevance, such as BIRC3, NOTCH1, SF3B1, MYD88, ATM, FBXW7, POT1, NFKBIE, CHD2, RPS15, IKZF3, ZNF292, ZMYM3, ARID1A, and PTPN11 [5, 19–22].

Richter transformation is defined as a transformation of CLL into aggressive lymphoma, most commonly diffuse large B-cell lymphoma (DLBCL). These patients typically have a poor response to traditional chemotherapy than de novo DLBCL. Moreover, their pattern of mutations is different from that of DLBLC-not otherwise specified (NOS). The risk of Richter transformation has been associated with prior therapy, U-CLL, NOTCH1 mutations, del(17p) and del(11q) [23–25].

Mutations in BTK, PLCG2, and CARD11 have been associated with resistance to BTK inhibitors [26, 27], whereas mutations in BCL2 have been linked to venetoclax resistance [28]. Therefore, choosing between immunochemotherapy and targeted therapies for CLL heavily depends on 17p/TP53 and IGHV status. Even though the treatment choice does not strictly require sequencing data, these data should be integrated into the decision and follow-up. As mentioned above, several genetic aberrations negatively impact prognosis, seem to confer a poorer response to conventional chemotherapy, and could be helpful when considering other treatment options based on BTK or BCL2 inhibition.

An ideal NGS panel for CLL should integrate the detection of IGHV SHM, gene mutations and CNAs (Table 1 and Fig. 1).

Table 1.

Summary of genetic alterations in B-cell lymphomas and their clinical utility

B-cell Lymphoma		Genetic alterations	Clinical significance	Refs.
Chronic lymphocytic leukemia (CLL)		IGVH SHM, Del(17p)/TP53, Del(11q), Del(13q), trisomy 12, NOTCH1, SF3B1, ATM, BIRC3, NFKBIE, EGR2, MYD88, XPO1, CHD2	DIAGNOSTIC/PROGNOSTIC	[5, 15, 17, 19, 22]
		BTK, PLCG2, BCL2, TP53, CARD11	PREDICTIVE	[26–28]
Lymphoplasmacytic lymphoma (LPL)		MYD88L265P#, CXCR4	DIAGNOSTIC/PROGNOSTIC/PREDICTIVE	[29, 32, 33]
Marginal zone lymphoma (MZL)	Spleenic MZL	KLF2, NOTCH2, TP53, NOTCH1, MLL2, ARID1A, SIN3A, TNFAIP3, MYD88, CARD11, trisomies 3, 18, Del(7q)	DIAGNOSTIC	[35, 37, 40]
	Nodal MZL	KLF2, NOTCH2, KMT2D, PTPRP, trisomies 3, 18, 7, 12	DIAGNOSTIC	[38, 41]
	Extranodal MZL	trisomies 3, 18, 12, T(11;18)#, T(1;14), T(3;14), T(14;18) (IGH::MALT1); TNFAIP3, CD79A, CD79B, CARD11, BIRC3, TRAF3, TNFRSF11A	DIAGNOSTIC/PROGNOSTIC (m7-FLIPI)/THERAPY	[42–45]
Follicular lymphoma (FL)	FL	T(14;18); KMT2D, EZH2*#, CREBBP#, EP300#, MEF2B#, ARID1A#, FOXO1#, CARD11#	DIAGNOSIS/PROGNOSTIC#/THERAPY*	[54, 56]
		NOTCH2, DTX1, UBE2A, HIST1H1E, MYC, TP53, CCND3, GNA13, S1PR2, P2RY8, POU2AF1, CDKN2A/B loss	PROGNOSTIC (HT)	[58–61]
	Diffuse FL	TNFRSF14, STAT6, CREBBP, EZH2	DIAGNOSTIC	[47]
	Pediatric TFL	TNFRSF14, MAP2K1	DIAGNOSTIC	[50]
	Duodenal type FL	TNFRSF14, CREBBP, EZH2	DIAGNOSTIC	[49]
Mantle cell lymphoma (MCL)		T(11;14), CCND2 and CCND3 rearr	DIAGNOSTIC	[1, 2]
		IGVH SHM, Del(17p)/TP53, ATM, NOTCH1/2, KMT2D	PROGNOSTIC	[64]
		Del(17p)/TP53, BIRC3, TRAF2, NSD2, CARD11	PREDICTIVE	[66–68]
Diffuse large B-cell lymphoma, not otherwise specified (DLBCL-NOS)		NOTCH2, BCL10, TNFAIP3, UBE2A, CD70, CCND3, DTX1, BCL2, EZH2, CREBBP, TNFRSF14, KMT2D, IRF8, EP300, GNA13, MYD88, CD79B, PIM1, PIM2, PRDM1, BTG1, CD58, NOTCH1, SGK1, SOCS1, TET2, STAT3, TP53; MYC rear., BCL2 rearr., BCL6 rear	DIAGNOSTIC/PROGNOSTIC	[73–77]
Large B-cell lymphoma with IRF4 rearrangement		IRF4 rear, IRF4, CARD11, MYD88, CD79B	DIAGNOSTIC	[69]
Large B-cell lymphoma with 11q aberration		11q aberration, GNA13	DIAGNOSTIC	[83]
High grade B-cell lymphoma (HGBCL)	HGBCL-DH-BCL2	MYC rear., BCL2 rear., BCL2, KMT2D, CREBBP, TNFRS14, EZH2	DIAGNOSTIC/PROGNOSTIC	[1, 2]
	HGBCL-DH-BCL6	MYC rear., BCL6 rear	DIAGNOSTIC	[1, 2]
	HGBCL-NOS	MYD88, CD79B, TBL1XR1, TP53, KMT2D	DIAGNOSTIC	[1, 2]
Burkitt Lymphoma (BL)	EBV +	MYC rear., MYC aSHM, CDKN2A, DDX3X	DIAGNOSTIC	[100, 101]
		TP53	PROGNOSTIC	[102]
	EBV -	MYC rear., TCF3, ID3, CDKN2A, DDX3X	DIAGNOSTIC	[100, 101]
		TP53	PROGNOSTIC	[102]
Hodgkin Lymphoma (HL)		XPO1, EP300, CREBBP, TP53, B2M, NFKBIE, TNFAIP3, STAT3, STAT6, PTPN1, ITPKB, GNA13, ARID1A, KTM2D, IGLL5, CSFR2B, BTK	FUTURE	[108–110]

The most relevant alterations are indicated in bold

Del deletion; SHM somatic hypermutation; EBV Epstein–Barr virus; DH double hit; NOS not otherwise specified

HT: Histological transformation; rear.: rearrangement. Refs.: references. The most relevant alterations are indicated in bold

Fig. 1.

Mature B-cell malignancies: cell of origin and main genetic alterations. Schematic representation of B-cell maturation process throughout the germinal center from naïve B-cells to memory and plasma B-cells, and their derived mature B-cell lymphoma. The recently proposed genetic subtypes for DLBCL are represented in the lower part of the figure, with the genes defining each subtype and the relationship with the DLBCL COO classification. The most relevant mutated genes are indicated in the box associated with each subtype. U-unmutated; M-mutated; CLL chronic lymphocytic leukemia; MCL mantle cell lymphoma; FL follicular lymphoma; MZL marginal zone lymphoma; DLBLC diffuse large B cell lymphoma; BL Burkitt’s lymphoma; DH Double hit; HL Hodgkin lymphoma; LPL/WM lymphoplasmacytic lymphoma/Waldenström macroglobulinemia; FDC follicular dendritic cells; COO cell-of-origin; GCB germinal center B cell; ABC activated B cell

Lymphoplasmacytic lymphoma/Waldenström macroglobulinemia

Lymphoplasmacytic lymphoma/Waldenström macroglobulinemia (LPL/WM) is an uncommon low-grade B-cell lymphoma, with an annual incidence of 3–4 cases per million people, characterized by bone marrow infiltration of clonal lymphoplasmacytic cells and the hypersecretion of immunoglobulin M [1, 2]. The lack of specific morphological, immunophenotypic, or chromosomal features requires the diagnosis to be made after excluding other small B-cell lymphomas.

Despite its rarity, our understanding of the biology of this disease has improved significantly in recent years with the identification of recurrent mutations in the MYD88 and CXCR4 genes [29]. More than 90% of LPL/MW cases bear the MYD88^L265P mutation. However, this is neither necessary nor specific for a diagnosis and can be seen in other B-cell lymphomas such as non-germinal center (GC) DLBCL, CLL, splenic marginal zone lymphomas, primary cutaneous DLBCL, leg type DLBCL, primary central nervous system DLBCL, or testicular DLBCL [30]. However, the co-occurrence of both mutations is highly suggestive of LPL/WM [31]. MYD88^L265P triggers tumor-cell growth through BTK, a target of ibrutinib [32]. CXCR4 is mutated in 30% of patients with LPL/WM, is associated with shorter treatment-free survival, and confers resistance to ibrutinib [32–34]. Patients with MYD88 and CXCR4 mutations have important differences in disease presentation and response to therapy with ibrutinib: overall response rates are 100% for those with MYD88^L265P and without CXCR4, 85.7% for patients with both mutations and 71.4% for those without any mutation [32].

Other reported genetic alterations include deletion of 6q (40–60% of cases), mutations in PRDM2 and BTG1 (approximately 90% of cases), HIVEP2, MKLN1, PLEKHG1, LYN, ARID1B, FOXP1, and ARID1A [32]. Molecular mutational diagnostic approaches, especially NGS, have helped refine the diagnostic criteria for LPL/WM. Molecular testing can be used for risk stratification and treatment planning of LPL/WM, beyond diagnostic purposes (Table 1 and Fig. 1).

Marginal zone lymphoma

Marginal zone lymphomas (MZLs) constitute a group of indolent B-cell lymphomas that arise from B lymphocytes in the marginal zone. They mainly compromise splenic MZL (SMZLs), with or without villous lymphocytes, nodal MZLs (NMZLs), and extranodal MZLs of mucosal-associated lymphoid tissue (EMZLs or MALTs). These are distinct clinical entities with specific diagnostic criteria and different genetic features, clinical behavior, and therapeutic implications [1, 2, 35]. MZLs account for approximately 10% of all NHLs derived from post-GC memory B-cells in the marginal zone of lymph nodes. The genomic landscape of MZL has been studied, revealing considerable overlap between mutated genes across distinct entities [1, 2, 36] (Table 1 and Fig. 1).

The most widely mutated genes in SMZL are KLF2 and NOTCH2. Other frequent mutations affect TP53 and NOTCH1. Other mutations leading to NF-κB pathway activation involving TNFAIP3, CARD11, MYD88 (outside the p.265 hotspot), or TRAF3 have been reported alongside alterations in chromatin remodelers such as KMT2D, ARID1A, and SIN3A [37, 38]. More than 30% of splenic MZLs are characterized by the deletion of 7q31–32 [36, 37, 39, 40].

NMZL and SMZL have several common alterations. Trisomies of chromosomes 3 and 18 co-occur in around 25% of SMZL and NMZL cases. The mutational profile of NMZL shows other recurrent clonal abnormalities like trisomies 7 and 12, and 6q deletion. Mutational analysis has identified mutations in KLF2, PTPRD, KMT2D, NOTCH2, LRP1B, TET2, and TNFRSF14. Other less frequent genetic alterations in NMZL include mutations in BRAF, EZH2, and HIST1H1E [38, 41].

The pathogenesis of EMZL/MALT is linked to several recurrent chromosomal aberrations, such as trisomies of 3, 12, and 18, which are found in 20–30% of cases. EMZL also presents recurrent chromosomal translocations. The most common is t(11;18)(p21;q21), which results in a functional chimeric fusion product between BIRC3 and MALT1 [36, 42]. It is associated with a low probability of response to antibiotics, Helicobacter pylori-negative, more advanced disease, and a lower risk of transforming to DLBCL. This translocation is specific for EMZL, since it has not been reported in SMZL or NMZL. Other common translocations are t(3;14)(p14;q32) (FOXP1::IGH), which is found in around 10% of EMZL cases; t(14;18)(q32;q21) (IGH::MALT1), which is present in 15–20% non-gastrointestinal EMZLs; and t(1;14)(p22;q32) (BCL10::IGH), which is a rare translocation found in 1–2% of EMZLs [1, 2, 43]. Homozygous deletion of the chromosomal band 6q23, involving TNFAIP3 (A20), has been described in EMZL and may contribute to lymphomagenesis by inducing constitutive NF-κB activation [44]. MYD88 mutation is detected in ocular adnexal MALT lymphoma (5% of cases) and can activate NF-κB, STAT3, and AP1 transcription factors [43]. Apart from the mutations of TNFAIP3 and MYD88, other alterations in NF-κB regulators have been identified (CD79A, CD79B, CARD11, BIRC3, TRAF3, and TNFRSF11A) [43]. The result of the recurrent genetic alterations mentioned above in the activation of the NF-κB activation pathway represents a possible therapeutic target for MALT lymphomas [45].

Follicular lymphoma

Follicular lymphoma (FL) is a malignancy derived from GC B-cells and the most common indolent B-cell lymphoma. It remains an incurable malignancy, but OS may last 20 years [46]. A key hallmark of FL is the t(14;18)(q32;q21) IGH::BCL2 translocation, the first hit in its oncogenesis. Recently, the new WHO and ICC classifications, recognized unique FL entities, such as in situ follicular B-cell neoplasm, duodenal-type FL, primary cutaneous follicle cell lymphoma, pediatric-type FL, and testicular FL.

A distinctive diffuse follicular lymphoma (dFL) variant lacking t(14;18) was first described in 2009 [47]. In a recent study, NGS analysis identified two molecular clusters: one was characterized by TNFRSF14 mutations, and the other showed few genetic alterations, a subgroup with STAT6 mutations concurrent with CREBBP mutations without TNFRSF14 and EZH2 mutations [48]. These findings suggest dFL might represent a subtype of t(14;18)-negative FL.

Gastrointestinal FL, especially duodenal-type FL (DTFL), frequently occurs as extranodal FL. This lymphoma is commonly found in the second part of the duodenum and exhibits indolent clinical behavior. It is morphologically and immunophenotypically indistinguishable from typical FL. More widespread use of NGS has identified that the mutation frequencies of recurrently mutated genes, including TNFRSF14, CREBBP, and EZH2, were not significantly different from typical FL, but KMT2D was less commonly mutated in DTFL [49].

Pediatric-type FL (PTFL) occurs in younger patients and shows a preference for the head and neck. Some studies have used NGS technologies to describe a specific mutational profile in PTFL distinct from those of other lymphomas, including typical FL. TNFRSF14 and MAP2K1 are the genes most frequently reported to be mutated in PTFL. One or the other is present in about 80% of cases, but they do not usually co-occur. This finding indicates that both genes are essential for the pathogenesis of PTFL [50].

In 2011, Morin et al. described frequent mutations of KMT2D and other chromatin-modifying genes (CMGs)[51]. These genes code for histone methyltransferases (EZH2, KMT2D) or histone acetylases (MEF2B, CREBBP, EP300) [30]. Alterations in these genes have been established as a central genetic hallmark of FL and are critical for determining GC and post-GC B-cell fate. The characteristic distinguishing FL from other B-cell lymphomas is the high rate of mutations in CMGs. The KMT2D gene is the most recurrently mutated CMG in FL (72%), followed by CREBBP (∼65% of FLs) and EP300 (15%) [52, 53]. Mutations of EZH2 are found in 25% of FLs; it has prognostic relevance and is currently investigated as a druggable target of therapeutic potential. EZH2 is a regulator of the GC phenotype, and mutations in this gene block B-cells and stop their differentiation into plasma cells.

Therapeutic targeting of epigenetic deregulation is an attractive concept. However, the most common CMG mutations (KMT2D and CREBBP) are loss of function/loss of protein events, which are difficult to target with drugs. This constraint has led researchers to focus on EZH2 mutations, with various companies developing inhibitors for EZH2 [54]. In a recent phase II study, tazemetostat, an oral inhibitor of EZH2, showed anti-tumor activity in patients with relapsed or refractory FL. Patients with or without mutations in EZH2, received tazemetostat and objective responses were observed in 69% of patients with mutated EZH2, and 35% of patients with wild-type EZH2 [55].

The Follicular Lymphoma International Prognostic Index (FLIPI) is the most widely used risk predictor. For failure-free survival, Pastore et al. proposed a clinicogenetic risk model, the m7-FLIPI score, which included the mutational status of seven genes (ARID1A, EZH2, EP300, FOXO1, MEF2B, CREBBP and CARD11), the Eastern Cooperative Oncology Group (ECOG) performance status and FLIPI [56]. m7-FLIPI defined a high-risk group with a 5-year failure-free survival rate of 38%, compared with 77% for the low-risk group, in patients who received first-line treatment with a combination of rituximab and chemotherapy (CVP or CHOP). But low-risk m7-FLIPI does not indicate a more indolent disease course, as all patients required chemotherapy. However, several studies concluded that the prognostic value of the m7-FLIPI clinicogenetic model seems to be dependent on the therapeutic regimen [57] (Table 1 and Fig. 1).

Histological transformation (HT) is reported to occur in 15–30% of patients with FL. HT refers to the evolution of an FL to a clinically aggressive lymphoma, such as DLBCL or Burkitt lymphoma, which is usually associated with poor prognosis and chemotherapy resistance. HT has been associated with alterations deregulating cell-cycle progression and DNA-damage responses (CDKN2A/B, MYC, and TP53) [58, 59], or other genes more commonly mutated in transformed samples than in FL tumors (CCND3, GNA13, S1PR2, and P2RY8) [60]. A study of FL samples from patients who did or did not transform discovered that the presence of mutations in four genes (NOTCH2, DTX1, UBE2A, and HIST1H1E) [61] was associated with a shorter time to transformation when mutated in the FL biopsies at diagnosis. This study also identified mutated genes enriched at transformation, like POU2AF1, which has roles in GC architecture and migration [61].

Mantle cell lymphoma

Mantle cell lymphoma (MCL) accounts for about 6% of NHL cases. The disease has two clinical presentations: common conventional MCL (cMCL) (90% of patients), which usually has an aggressive clinical course (SOX-11-positive cells and an unmutated IGHV), and an indolent clinical presentation (10% of patients), which generally presents as non-nodal leukemic phase (nnMCL) (SOX-11-negative, mutations of CCND1 and TLR2, and somatic hypermutation of IGHV) (2,3). MCL is typically an aggressive and incurable B-cell malignancy, but some patients may follow an indolent clinical course.

MCL is characterized by the t(11;14)(q13;q32) translocation, leading to the overexpression of Cyclin D1 (CCND1), which is detected in nearly 95% of cases. Although the detection of CCND1 helps support an MCL diagnosis in unclear mature B-cell neoplasms, FISH is the current gold standard assay used to identify recurrent cytogenetic alterations, although this methodology may not detect complex or cryptic rearrangements [62, 63]. The few cases without this characteristic CCDN1 rearrangement are characterized by CCND2 or CCND3 translocations [1].

Some studies have identified significantly mutated genes such as ATM, and the tumor suppressor TP53, but also found NOTCH2 mutations in aggressive tumors with a worse prognosis [64].

MIPI (MCL International Prognostic Index) is based on a weighted sum of performance status, age, lactate dehydrogenase (LDH) levels and white blood cell count. Additional modifications, such as “MIPI genetic” (MIPIg), are being explored to refine this score. MIPIg is associated with an increased risk of progression and death when mutations of KMT2D and deletion or mutation of TP53 are present [65]. At diagnosis, the frequency of TP53 mutations is about 11–25% but increases to 45% at relapse. TP53 deletion (determined by FISH) and TP53 mutations were associated with the worst survival [66] (Table 1 and Fig. 1).

Ibrutinib-refractory MCL patients exhibit poor survival and lack an optimal management strategy. Some studies have investigated the relationship between mutations in BIRC3, TRAF2, and CARD11 genes, MCL progression, and ibrutinib resistance [67]. Recently, a study explored the mutational profile in a subset of patients who developed disease progression or disease transformation on ibrutinib treatment. Using targeted NGS, they detected TP53 alterations in 75% of patients after progression on ibrutinib. They found mutations in chromatin-modifier genes, such as NSD2 in 75% of patients with transformed MCL on ibrutinib therapy. They concluded that NSD2 mutations are involved in altered methylation and chromatin dysfunction, leading to aberrant gene expression with pathological significance in MCL progression and ibrutinib resistance [68].

Diffuse large B-cell lymphoma

DLBCL is the most common subtype of NHL (30–35% of cases), characterized by large mature B-cell morphology and phenotype [1, 2]. DLBCL is heterogeneous in its clinical behavior and pathological and molecular diagnosis. The most common type of DLBCL (80%) is DLBCL- NOS, which has no specific clinical presentation or pathology. Three molecular subtypes have been recognized by the WHO classification based on the cell of origin (COO): germinal center (GCB), activated B-cell (ABC) or non-GCB, and unclassifiable DLBCL [1, 69]. This classification has a prognostic value, with ABC-DLBCL associated with poorer outcomes [70]. However, the COO does not imply different treatments, R-CHOP (rituximab, cyclophosphamide, adriamycin, vincristine and prednisone) being the backbone of therapy for all subtypes [71]. Up to 40% of DLBCLs will be refractory to first-line treatment or will recur [72].

Several researchers have recently proposed new genetic groups with broad concordance, suggesting that mutational analysis could be a promising alternative to classify DLCBL with prognostic and theragnostic values [1, 2, 73–76]. Each genetic cluster harbors a distinct mutational profile. Schmitz et al. defined the following subtypes: MCD (co-occurrence of MYD88^L265P and CD79B mutations), BN2 (with BCL6 fusions and NOTCH2 mutations), N1 (NOTCH1 mutations), EZB (EZH2 mutations and BCL2 translocations), A53 (TP53 mutations and deletions) and ST2 (SGK1 and TET2 mutations) [73, 74]. In parallel, another approach distinguished five subsets of DLBCL, including two ABC-DLBCL groups, one with low risk and a possible marginal zone origin (C1), and the other a high-risk group (C5) enriched in cases with mutations in MYD88, CD79B, and PIM1; two subsets of GC-DLBCLs with favorable (C4) and poor (C3) outcomes, and an ABC/GC-independent group (C2) with biallelic inactivation of TP53, CDKN2A loss, and associated genomic instability [75]. To unify both classifiers, Lacy et al. established the subtypes NOTCH2, MYD88, BCL2, TET2/SGK1, and SOCS1/SGK1, according to the mutated genes that are most highly enriched in each one [76]. However, all these studies are mainly based on whole-genome sequencing and are not applicable in routine clinical practice. Hence, some efforts have been made to propose easy and feasible classifiers built with a combination of mutational and translocation data of a selected set of genes representing each genetic subtype [77, 78].

A consensus may soon be reached to develop an NGS panel for assigning genetic groups (Tables 1 and 2, and Fig. 1).). To move towards personalized medicine, it should include at least the analysis of the MYC, BCL2, and BCL6 rearrangements, COO determination at diagnosis, and the chosen NGS panel. Although there is no agreement about the classification of the genetic subtypes, an adequate preliminary NGS panel could be designed, including a set of common genes from the genetic classifiers: MYD88, CD79B, PIM1, PIM2, PRDM1, BTG1, CD58, ETV6, and TBL1XR1 for the MCD subtype; NOTCH2, TNFAIP3, BCL10, UBE2A, CD70, CCND3, and DTX1 for the BN2 subtype; EZH2, CREBBP, TNFRSF14, KMT2D, BCL2, IRF8, and EP300 for the EZB subtype; SGK1, SOCS1, TET2, NFKBIA, and STAT3 for the ST2 subtype, and NOTCH1 for the N1 subtype [77, 78].

Table 2.

DLBCL genetic subtypes

Genetic types	Subtypes	Main alterations	COO	Clinical outcome [73–77]
MCD/C5/MYD88		MYD88L265P, CD79B, PIM1, PRMD1, BTG1. CD58	ABC	Bad prognosis
BN2/C1/NOTCH2		BCL6 rear., NOTCH2, BCL10, TNFAIP3, CD70, SPEN, CCND3, UBE2A, DTX1	ABC / GCB	Intermediate prognosis
EZB/C3/BCL2	EZB	BCL2 rear., BCL2, EZH2, CREBBP, IRF8, KMT2D, TNFSR14, EP300. MYC rear	GCB	Good prognosis
	EZB-MYC		DH	Bad prognosis
N1/NEC		NOTCH1	Mostly ABC	Bad prognosis
A53/C2/NEC		TP53, aneuploidy	Mostly ABC	Intermediate prognosis
ST2/C4		SOCS1, TET2 & SGK1, STAT3, NFKBIE, BRAF, CD83	Mostly GCB	Good prognosis

Integration of the genetic subtypes described by Wright [74], Chapuy [75], and Lacy [76] and their relation to the cell-or-origin and patient clinical outcome

COO cell-of-origin; ABC activated B-cell; GCB germinal center B-cell; DH double hit; rear. rearrangement

Extranodal DLBCL cases arising in immune-privileged sites, such as primary DLBCL of the central nervous system (PCNSL), primary DLBCL of the testis, primary cutaneous DLBCL, leg type, and other related entities, such as breast DLBCL [79], have similar molecular features and there are some recent controversies about their classification [1, 2]. Most of the lymphomas in these locations are non-GCB/ABC type and seem to share common molecular features such as the high prevalence of MYD88^L265P and CD79B mutations that characterize the DLBCL-MCD/C5/MYD88 genetic subtype.

DLBCL associated with viral agents (EBV-associated, HHV-8-associated) are rare and, in most instances, their diagnosis is based on the clinical presentation and pathological features; a set of mutated genes possibly specific for this disease has been described in EBV-positive DLBCL, including CCR6, CCR7, DAPK1, TNFRSF21, CSNK2B and YY1 [80], as well as common PD-L1 genetic aberrations [81]

Large B-cell lymphoma (LBCL) with the IRF4 rearrangement occurs most commonly in younger patients [1, 2]. It prefers the lymphoid tissue of Waldeyer’s ring and head and neck lymph nodes [82]. In a study with pediatric LBCL, the subtype with IRF4 rearrangement had a GCB phenotype with frequent mutations in IRF4 and NF-κB pathway genes (CARD11, MYD88, CD79B) [69].

LBCL with 11q aberration [1] (named “HGBCL with 11q aberrations” in the 5th WHO classification) [2] was classified as Burkitt-like lymphoma with 11q aberration in the previous WHO classification. The mutational pattern is more like that of GCB-DLBCL than Burkitt-like, lacks MYC rearrangements, and shows frequent mutations in GNA13. Genomic alterations affecting the ID3-TCF3 complex are quite exceptional in LBCL-11q [83].

High-grade B-cell lymphoma

The recently revised 5th edition of the WHO Classification of Hematolymphoid Tumors and the ICC now recognize the following groups: high-grade B-cell lymphoma (HGBCL) with MYC and BCL2 rearrangements (with or without BCL6 rearrangement, HGBCL-DH-BCL2) [1, 2], a provisional entity with MYC and BCL6 rearrangements (HGBCL-DH-BCL6) [1] and HGBCL-NOS.

The mutational findings in the HGBCL-DH-BCL2 lymphoma subtype are relatively homogeneous and similar to those of FL, harboring frequent molecular abnormalities in BCL2, KMT2D, CREBBP, TNFRS14, and EZH2. Several recent studies described DLBCL cases with a gene expression signature (GEP) similar to that of HGBCL (DH-like GEP signature) [84–86] and have frequent mutations in MYC, BCL2, DDX3X, TP53, and KMT2D [85] (Table 1 and Fig. 1).

Lymphoid neoplasms with dual MYC and BCL6 rearrangements are now considered a subtype of DLBCLNOS, or HGBCL, and their gene expression profiles are highly heterogeneous and unrelated to those of DLBCL/HGB-cell lymphomas with MYC and BCL2 rearrangements.

HGBCL-NOS is recognized as another subtype and covers cases that cannot be included in the other entities. From a molecular perspective, it is a heterogeneous category that includes activated B-cell lymphomas with mutations of MYD88, CD79B, or TBL1XR1. The most frequent mutations are found in TP53 and KMT2D. Gene-expression profiling showed that approximately half the cases of HGBCL-NOS harbor the previously described DH-like GEP signature [85–87].

Burkitt’s lymphoma

Burkitt’s lymphoma (BL) is a highly aggressive mature B-cell NHL characterized by rapid proliferation [1, 2, 88]. It accounts for 1–2% of adult lymphomas, whereas it is a common pediatric neoplasm [89].

Historically, three BL clinical variants have been described and are currently recognized by the WHO classification: endemic, non-endemic or sporadic, and immunodeficiency associated. The three subtypes present identical morphological and immunophenotypical features but are clinically and epidemiologically different [1, 2, 90]. All clinical variants share pathological features. A defining feature of BL is the constitutive overexpression of MYC due to a translocation of this oncogene along with one of the three immunoglobulin genes located on chromosomes 14, 2 and 22, accounting for nearly 80%, 15%, and 5% of cases, respectively. Other typical lymphoma translocations, such as BCL2 and BCL6, do not occur in this entity [2, 88]. However, MYC deregulation is not sufficient for lymphomagenesis. Therefore, other genetic alterations are commonly detected by next-generation techniques [91, 92].

Distinguishing BL from other HGBCLs is an important challenge in clinical practice that must be addressed. Gene expression profile studies have revealed a distinct BL signature, making it an independent entity [93–95]. Examining the BL genetic landscape, several studies have demonstrated that ID3/TCF3-dependent centroblast gene expression program, tonic PI3K-AKT-mTOR signaling, and cell cycle deregulation and apoptosis are essential mechanisms for BL lymphomagenesis. Inactivating mutations in tumor-suppressor genes, such as TP53, CDKN2A, and DDX3X, are recurrent in BL [92, 96, 97]. Moreover, TP53 inactivation has been reported as a potential prognostic biomarker because these mutations were enriched in refractory patients [96]. Likewise, the higher prevalence of BL among males may be partially accounted for by inactivating alterations in DDX3X. Loss-of-function DDX3X mutations moderate MYC-driven global protein synthesis, and established malignant cells restore full protein synthetic capacity by aberrant expression of their Y chromosome homolog (DDX3Y), whose expression is usually restricted to the testis [97]. Cell migration and dissemination disturbances are partly due to the inactivation of P2RY8 and GNA13 [98], and high levels of proliferation are due not only to CCND3 activation but also to CDKN2A inactivation [92]. Another oncogenic mechanism is the constitutive activation of BCR signaling by mutations in the transcription factor TCF3 and its negative regulator ID3.

Consequently, BCR signaling activates PI3K-AKT-mTOR signaling, highlighting alterations targeting PTEN and FOXO1, essential for BL survival [92, 99]. Finally, although various chromatin regulators are recurrently mutated in BL, mutations in EZH2, CREBBP, and KMT2D are rarely observed, unlike in GCB-DLBCL [100]. Furthermore, studies searching for new BL subdivision strategies with clinical implications suggest that EBV status could be an up-and-coming approach [100, 101]. EBV-positive BL shows significantly higher levels of aberrant SHM, fewer driver mutations, particularly in the apoptosis pathway, and fewer mutations in TCF3 and ID3 [2, 100] (Table 1 and Fig. 1).

In fact, the new WHO classification recommends that EBV-positive and EBV-negative BL be recognized as distinct entities [2]. Regarding differential diagnosis, identification of MYC translocations for BL diagnosis and the absence of the typical chromosome 11q-gain/loss pattern observed in Burkitt-like lymphoma are mandatory. As for NGS, it would be recommendable to study differential mutational patterns in TCF3 and ID3, which are recurrently mutated in BL, and EZH2, CREBBP, and KMT2D, which are mutated in other lymphomas. Moreover, TP53 mutational status could provide prognostic information [102].

Hodgkin lymphoma

Hodgkin lymphoma (HL) is a clonal neoplasm derived from B-cells in most cases. HL constitutes 25–30% of all lymphomas and is subdivided into classic Hodgkin lymphoma (cHL) (95% of cases) and nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) (5% of cases). Unlike other tumors, neoplastic cells (Hodgkin and Reed–Sternberg [HRS] and lymphocytic and histiocytic [L&H] cells, in the case of NLHPLN) account for less than 1–2% of the total cellularity. The ICC and new WHO classifications propose “nodular lymphocyte predominant B-cell lymphoma” (NLPBL) as a new term for NLPHL [1, 2], based on the significant biological and clinical differences from cHL and its close relationship to T-cell/histiocyte-rich large B-cell lymphoma [103].

cHL, primary mediastinal large B-cell lymphoma, and mediastinal gray-zone lymphomas are related diseases with common genetic alterations, phenotypes, and clinical features, including anterior mediastinal involvement [1]. cHL is a monoclonal proliferation of HRS cells (and their variants) accompanied by a reactive microenvironment, the recognition of both elements being essential for its diagnosis. Cytogenetically, besides aneuploidy and hyper-tetraploidy, HRS cells show recurrent chromosomal imbalances, including gains of 2p13 (REL), 9p24.1 (CD274 (PDL1), PDCD1LG2 (PDL2), JAK2), 17q21 (MAP3K14), and loss of 6q23-q24 (TNFAIP3) [104–107].

The scarcity of tumor cells has hampered the genetic characterization of cHL. However, genetic studies based on tissue microdissection or cell isolation by flow cytometry have shown the dysregulation of specific pathways rather than mutations in specific genes. Studies have revealed recurrent somatic mutations in NF-κB pathway components (TNFAIP3, NFKBIA, NFKBIA, REL); the JAK/STAT pathway (SOCS1, PTPN1, STAT6, STAT3, CSF2RB) [108–111]; epigenetic regulators such as EP300, CREBBP [108], and TP53; and regulators of immune escape such as inactivating mutations in the gene of the MHC class 1 component B2M, and the MHC class 2 transactivator (C2TA) [110, 112], and the FAS gene [113], which favor the evasion of apoptosis and cell proliferation. Other relevant signaling pathways in cHL pathogenesis include MAPK/ERK, AP1, PI3K/AKT, and NOTCH [114] (Table 1 and Fig. 1).

Mature T-cell and NK-cell lymphomas

Non-Hodgkin T-cell lymphomas (NHL-Ts) is a heterogeneous group of relatively rare malignancies with generally aggressive clinical behavior originating in T lymphocytes or natural killer (NK) cells. T-cell lymphomas (TCLs) account for 5–10% of all NHLs in Western countries, with an overall incidence of 0.5–2.0 per 100,000 inhabitants per year [115]. Depending on the histological subtype, NHL-T debuts with a nodal or extranodal presentation [116, 117]. The recent WHO and ICC classifications recognized the following subtypes as the primary nodal T and NK-derived neoplasias: follicular helper TCL (TFH); anaplastic large cell lymphoma (ALCL; ALK-positive and ALK-negative); peripheral TCL not otherwise specified (PTCL-NOS); and primary nodal EBV-positive T-/NK-cell lymphoma [1, 2, 118, 119]. Regarding extranodal subtypes, the most common extranodal entities TCL subtypes are the cutaneous TCL (CTCL), extranodal NK/T-cell lymphomas nasal type (ENKL), breast implant-associated anaplastic large cell lymphoma (BIA-ALCL), intestinal TCL (ITCL) and hepatosplenic TCL (HSTCL) [120] (Table 3 and Fig. 2).

Table 3.

Summary of genetic alterations in T-cell lymphomas and their clinical utility

T-cell Lymphoma		Genetic alterations	Clinical significance	Refs.
Follicular helper T-cell lymphoma (TFH)	Angioimmunoblastic (AITL)	IDH2R172, RHOAG17V, TET2, DNMT3A$, VAV1, CD28, ICOS, FYN and LCK rearr	DIAGNOSTIC/PREDICTIVE$	[121, 122]
	Follicular type
	Not otherwise specified (NOS)
Anaplastic large cell lymphoma (ALCL)	ALK-positive	ALK fusion, NOTCH1*	DIAGNOSTIC/THERAPY*	[123, 125]
	ALK-negative	DUSP22 rearr#, JAK1, JAK3, STAT3	DIAGNOSTIC/PROGNOSTIC#/PREDICTIVE	[127, 128]
		Del(17p)/TP53, TP63 rearr., PRDM1 loss	PROGNOSTIC	[128]
Peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS)	PTCL-TBX21	TET1, TET3, DNMT3A	DIAGNOSTIC/PROGNOSTIC#	[126, 130]
	PTCL-GATA3	TP53#, PRDM1, CDKN2A/B, RB1 and PTEN loss, STAT3 and MYC gain		[120, 129, 130]
Primary nodal EBV-positive T-/NK-cell lymphoma		TET2, PI3KCD, STAT3, TP53, CARD11	DIAGNOSTIC	[131, 132]
Cutaneous T-cell lymphoma (CTCL)		PLCG1, NFATC2, NFAT5, ZEB1, PRKCQ, RHOA, VAV1, PREX2, CTCF, ARID1A, TRRAP	DIAGNOSTIC	[134, 136, 137, 139]
Hepatosplenic T-cell lymphoma (HTCL)		STAT5B, STAT3, PIK3CD, SETD2, IN080, ARID1; Loss of 7p, amplification of 7q	DIAGNOSTIC	[140, 141]
Breast implant-associated ALCL		STAT3, JAK1, JAK3, DNMT3, TP53	DIAGNOSTIC	[143–145]
Extranodal NK/T-cell lymphoma (ENKT), nasal type		TP53#, DDX3X#, Del(6q), STAT3, JAK3, STAT5B	PROGNOSTIC#	[119]

The most relevant alterations are indicated in bold

Del deletion; EBV Epstein–Barr virus; NK natural killer; rear: rearrangement. Refs. references

*clinical significance

Fig. 2.

Essential genetic alterations in T/NK-cell malignancies. Schematic representation of T-cell and NK-cell derived malignancies, including the lymphocyte subtype they are derived from, and their pathological location in nodal or extranodal sites. The main genetic alterations of each entity are indicated in the box associated with each subtype. TFH follicular helper T-cell; FDC follicular dendritic cells; AITL angioimmunoblastic lymphoma; NOS not otherwise specified; PTCL peripheral T-cell lymphoma; CTCL cutaneous T-cell lymphoma; HTCL hepatosplenic T-cell lymphoma; EATL enteropathy-associated T-cell lymphoma; MEITL monomorphic epitheliotropic intestinal T-cell lymphoma; BIA-ALCL breast implant-associated ALCL; ENKT extranodal NK/T-cell lymphoma

Nodal T- and NK-cell lymphomas

Follicular helper T‑cell lymphoma

Follicular helper T-cell (TFH) lymphoma comprises three entities with shared molecular features and a gene expression signature similar to TFH cells that define this subtype: the angioimmunoblastic (AITL), follicular, and NOS types. These lymphomas have similar mutational landscapes, including loss of function mutations in the methylation-associated genes TET2 (in around 80% of cases) and DNMT3A (30–40%). Other recurrently mutated genes are CD28, RHOA (G17V), and IDH2 (R172), primarily seen in a subset of AITLs, and genes in the TCR signaling pathway. ICOS::CD28, ITK::SYK and fusions involving VAV1 are some of the recurrent fusions that have been described. In summary, TFH lymphomas share alterations in epigenetics and TCR signaling genes [118, 121, 122].

Anaplastic large cell lymphoma

Anaplastic large cell lymphoma (ALCL) is characterized by pleomorphic tumor cells with uniform CD30 expression and includes four distinct subtypes [1, 2]: two nodal and two extranodal subtypes. This section will focus on systemic ALK-positive (with the ALK rearrangement, with several different fusion partners) and ALK-negative ALCL [118, 123, 124], which is more frequent in children and young adults. The most frequent ALK rearrangement is the t(2;5)(p23;q35), which leads to the fusion of nucleophosmin (NPM1) to ALK, resulting in a chimeric protein. These cases also have recurrent mutations in NOTCH1, which may represent a candidate therapy target [118, 124, 125]. Mutations in TP53 and epigenetic regulators (EP300, KMT2D/C) have also been reported [126]. Immunophenotypic, histological, molecular, and clinical data are needed to diagnose ALK-positive ALCL correctly.

ALK-negative ALCL is quite a heterogeneous entity. The DUSP22 rearrangement is present in around 20–30% of cases. Indeed, it has been defined as a distinct genetic ALK-negative subtype [127]. Around 60% of ALK-negative cases show activation of JAK-STAT3 through mutations mainly in JAK1, JAK3, and STAT3, or rearrangements involving TYK2, ROS1, and FRK. This activation is not present in ALK-negative ALCL with DUSP22 rearrangements. A small percentage of ALK-negative ALCL cases feature the TP63 rearrangement and losses of TP53 and PRDM1, which are associated with an aggressive course [128].

Peripheral T-cell lymphomas, PTCL-NOS

This group accounts for 34% of nodal PTCLs [117], including cases that do not meet the criteria for inclusion in the other defined entities. FAT1 mutations have been reported to be frequent in PTCL-NOS and are associated with a worse prognosis [31, 129, 130]. This category is subdivided into two molecular subgroups, PTCL-TBX21 and PTCL-GATA3, which have different genetic landscapes [117]. It is characterized by high mortality and poor prognosis [31]. PTCL-GATA3 is more complex genetically and has a worse prognosis. It has frequent losses and mutations of TP53 and PRDM1, losses of CDKN2A/B, RB1, and PTEN and gains of STAT3 and MYC. PTCL-TBX21 shows frequent mutations in CMG, such as TET1, TET3, and DNMT3A.

Primary nodal EBV-positive T-/NK-cell lymphoma

Primary nodal EBV-positive T-/NK-cell lymphoma is labeled a new provisional entity in ICC and WHO classifications [1, 2]. It is a rare disease of elderly or immunocompromised patients with poor prognosis. This entity shows low genomic instability, downregulation of EBV microRNAs, and frequent mutations in STAT3, TET2, CARD11, BCOR, ARID1B, TP53, and PI3KCD genes [118, 131, 132].

Extranodal T- and NK-cell lymphomas

Cutaneous T-cell lymphoma

Cutaneous T-cell lymphoma (CTCL) is a subtype of ectopic lymphoproliferative disease originating in the skin. The most common clinical manifestation is mycosis fungoides (MF), in which patients show cutaneous patches, plaques, or tumors [133]. Histologically, a CD4 + lymphocyte infiltration is observed. Sézary syndrome (SS) is a leukemic form of MF. Whole-genome, exome and targeted sequencing for SS and MF have revealed genomic alterations, including somatic mutations and CNAs), in genes that are well-known participants in key cellular activities like DNA damage (e.g., TP53 (mut and del) and ATM (mainly del)), TCR signaling (PLCG1, ZEB1 (del)), NF-κB signaling (CARD11 and TNFRSF1B), CCR4/MAPK signaling (CCR4), JAK/STAT signaling (JAK1/2/3, STAT3, and STAT5B), cell migration (RHOA, the most common affecting p.N117, which is not found in other T-cell neoplasias, and VAV1), and chromatin remodeling (ARID1A and CTCF) [134–138]. Although most of them are not specific to CTCL diagnosis, some could be of theragnostic value, such as the PLCG1 or JAK/STAT genes, and RHOA (p.N117I) seems to be CTCL-specific [139].

T-cell lymphomas of the gastrointestinal tract

Intestinal T-cell lymphomas (ITCLs) comprise two main entities: enteropathy-associated TCL (EATL) and monomorphic epitheliotropic intestinal TCL (MEITL), as well as ITCL-NOS, diagnosed by exclusion and without specific clinicopathological characteristics (for a recent review see [120]).

EATL is more prevalent in Western populations and typically occurs in patients with celiac disease. It is characterized by recurring mutations in JAK/STAT pathway (JAK1 and STAT3), NFkB (TNFAIP3), and epigenetic regulators (KMT2D, BCOR and DDX3X).

MEITL is a rare ITCL, and it is unrelated to celiac disease, affects older patients and is the main ITCL in Asia. The most prevalent mutations occur in STAT5B and JAK3, and mutually exclusive alterations in BRAF, KRAS and NRAS are detected more frequently than in EATL. Mutations in the epigenetic gene SETD2 are also frequently detected in this entity.

Hepatosplenic T-cell lymphoma

Hepatosplenic T-cell lymphoma (HSTCL) is a rare and aggressive TCL that affects adolescents and young adults. This group represents less than 1% of NHLs and is related to chronic immunosuppression [140]. Isochromosome (7q) can usually be detected, making it the most frequent chromosomal abnormality. Loss of 7p and amplification of 7q result in the altered expression of several oncogenes located on chromosome 7 (CHN2, ABCB1, PPP1R9A) [141]. Recent studies have detected genetic alterations that could be considered oncogenic drivers in the near future. SETD2, IN080, and ARID1 mutations are involved in chromatin modification (occurring almost exclusively in HSTCL compared with other TCL subtypes). STAT5B, STAT3, and PIK3CD mutations have also been detected [140].

Breast implant-associated ALCL

Breast implant-associated ALCL (BIA-ALCL) is a rare TCL that develops after a relatively long period following breast implant placement (8–11 years). It appears in fluids and capsules around the prosthesis. Its clinical, genomic, and molecular characteristics differ from other ALCLs [1]. They show clonal TCR rearrangements, and STAT3 is recurrently mutated in up to 64% of BIA-ALCLs; other mutated genes include JAK1, JAK3, DNMT3A, and TP53 [142–145]. ALK, DUSP22, and TP63 rearrangements have not been found in BIA-ALCA cases [144, 146, 147].

Extranodal NK/T‑cell lymphoma, nasal type

Extranodal NK/T‑cell lymphoma (ENKTL), nasal type comprises around 20–25% of mature T/NK-cell lymphomas in Asia and Central and South America, but only 5% in Europe and North America. It usually affects the upper aerodigestive tract. There is a strong association of ENKTL with EBV infection, although the mechanisms of its role are not fully understood. Deletion 6q21-25 is one of the most frequent genomic abnormalities, including PRDM1, PTPRK, and FOXO3 genes. The most common mutations affect the JAK/STAT pathway (STAT3, JAK3, STAT5B), tumor suppressors genes (TP53, DDX3X), and epigenetic modifiers (TET2 [~ 5–10%], KMT2D, KMT2C) [1, 2, 119, 120].

Liquid biopsy in lymphomas

In recent years, high-throughput sequencing techniques have been successfully applied to the so-called “liquid biopsy” (LB), mainly to circulating tumor DNA (ctDNA). ctDNA represents the cell-free DNA released by the tumoral cell to body fluids, blood plasma being the most thoroughly analyzed form. The use of LBs, especially ctDNA, to support classical diagnostic tools in solid biopsy is increasing, mainly due to the advantages of this technique over the classic ones, as it is a minimally invasive approach that allows disease monitoring and detection of minimal residual disease (MRD) [148]. In fact, MRD detection in peripheral blood is a potent tool in various cancers, including hematological malignancies. Several studies have proved its clinical significance in DLBLC, cHL, MCL, PTCL [149–156], and, more recently, in indolent lymphomas, such as FL [157, 158].

Although the gold standard for mutational profiling is based on tissue biopsy, cfDNA genotyping would complement it. Moreover, it could be an excellent approach in certain situations, such as when dealing with inaccessible tumors or small tissue biopsies or when a re-biopsy is needed following relapse, transformation, or other clinical events. Concordance between ctDNA and FFPE in aggressive lymphomas is around 80% and somewhat lower in indolent lymphomas with low tumor burden [157]. However, even in these cases, ctDNA analysis has proved to be valid.

Basal or pre-treatment cfDNA analysis is valuable for tumor genotyping and as a surrogate for the tumor burden. The serial analysis allows the real-time follow-up of treatment response, clonal evolution monitoring, and MRD measurement [149–152, 157, 158]. For this reason, ctDNA genotyping may soon complement tissue analysis.

Additionally, the analysis of “in-phase” variants, done by two research groups who took slightly different approaches [159, 160], has improved the detection limit of ctDNA. This analysis tracks two or more somatic mutations in the same cfDNA fragment, lowering the background signal due to technical or biological errors. This approach is advantageous in lymphomas enriched in regions of aberrant SHM, leading to potentially more sensitive ctDNA detection and, thereby, greater MRD detection capacity [157, 159, 160].

However, efforts must still be made to standardize every step to successfully apply LB in the clinical milieu, from sample manipulation to bioinformatic analyses [148, 161, 162]. These technical and analytical considerations are serious challenges and are the focus of multiple cooperative efforts to allow its eventual application in routine clinical practice.

Conclusions

Thanks to the development of genomics, molecular data have enabled the better diagnosis of lymphomas, and these alterations are now part of the diagnostic criteria. NGS allows the simultaneous detection of multiple alterations, including mutations, copy number alterations and structural aberrations, and this genomic information may be combined with morphology and immunophenotyping for the purposes of diagnosis and prognosis. However, standardization throughout the entire process and quality controls are prerequisites for NGS implementation in lymphoma diagnosis. The panels proposed in this review (Table 4) for SNVs and CNAs detection are intended to be a helpful tool to encourage the implementation of NGS-based lymphoma diagnosis.

Table 4.

Proposed next-generation sequencing panels for B- and T-cell lymphomas

B-cell lymphomas							T-cell lymphomas
Mandatory		Recommended		CNAs			Mandatory		Recommended		CNAs
B2M	KTM2D	ARID1A	MEF2B	ATM	CN gain	11p	CD28	RHOA	ARID1A	NOTCH1	TP53	CN loss	17p13.1
BCL10	MYC	ARID1B	MKLN1	BCL2	CN gain	18q21.33	DNMT3A	SETD2	CTCF	PREX2	MYC	CN gain	8q24.21
BCL2	MYD88	ATM	NFKBIE	BCL6	CN gain	3q27.3	IDH2	STAT3	FYN	PRKCQ	CDKN2A	CN loss	9p21.3
BCL6	NOTCH1	CD79A	NSD2	CDKN2A	CN loss	9p21.3	JAK1	STAT5B	IN080	RB1
BIRC3	NOTCH2	CSFR2B	P2RY8	FOXO1	CN gain	13q	JAK3	TET1	LCK	TET3
BTG1	PIM1	DDX3X	PLEKHG1	MYC	CN gain	8q24.21	PI3KCD	TET2	NFAT5	TRRAP
BTK	PIM2	FBXW7	POT1	TP53	CN loss	17p13.1	PLCG1	TP53	NFATC2
CARD11	PLCG2	FOXO1	PTPN11	CCND1	CN gain	11q13	PRDM1	VAV1
CCND3	POU2AF1	HIVEP2	S1PR2	Del (7q)	CN loss	7q31–32
CD58	PRDM1	IGLL5	SIN3A	TRISOMY 3	CN gain
CD70	SF3B1	IKBK	RPS15	TRISOMY 12	CN gain
CD79B	SGK1	IKZF3	TRAF2	TRISOMY 18	CN gain
CHD2	SOCS1	ITPKB	TRAF3
CREBBP	STAT3	LYN	XPO1
DTX1	STAT6	MAP2K1	ZMYM3
EP300	TCF3	MAP3K14	ZNF292
EZH2	TET2
GNA13	TNFAIP3
HIST1H1E	TNFRSF14
IRF8	TP53
ID3	UBE2A
KLF2	IGHV

CNAs Copy number alterations; CN Copy number

Data availability

Not applicable.

Declarations

Conflict of interest

The authors declare no conflict of interest.

Ethical approval

Not applicable.

Informed consent

Not applicable.