Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) is a typical indolent non-Hodgkin lymphoma (NHL) characterized by accumulation of malignant B cells in bone marrow (BM), peripheral blood and lymph nodes. Richter’s syndrome refers to high-grade NHL or Hodgkin lymphoma (HL) in patients with CLL/SLL. Approximately 2%–8% of patients with CLL/SLL would develop diffuse large B cell lymphoma (DLBCL), and less than 1% would present classic HL (1). In rare cases, CLL/SLL patients present T-cell malignancies with an incidence of approximately 1%, which is usually in the anaplastic large cell type or with a cytotoxic phenotype (2). Monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL) is a rare primary intestinal T-cell lymphoma formerly known as type 2 enteropathy-associated T-cell lymphoma (EATL), resulting in very poor prognosis. In this study, we reported an extremely rare case presenting simultaneous CLL/SLL and MEITL.

A 63-year-old gentleman was referred to our hospital due to needle-like pain in the upper abdomen for at least 15 days on November 18, 2020. The pain was transient and was relieved shortly, but he showed serious pain and vomiting occasionally. Gastroscopy showed reflux esophagitis and chronic atrophic gastritis. Doppler ultrasonography of the abdomen showed no abnormalities. Routine blood examination results were as follows: white blood cell, 17.9 × 10⁹/L (normal range: 4–10 × 10⁹/L); lymphocyte ratio, 62.5% (normal range: 20%–40%); lymphocyte, 11.0 × 10⁹/L (normal range: 1–3 × 10⁹/L); erythrocyte, 4.7 × 10¹²/L (normal range: 3.5–5.5 × 10¹²/L); hemoglobin, 87 g/L (normal range: 110–160 g/L); mean corpuscular volume (MCV), 67 fL (normal range: 82–92 fl); corpuscular hemoglobin (MCH), 19 pg (normal range: 27–31 pg); mean corpuscular hemoglobin concentration (MCHC), 284 g/L (normal range: 320–360 g/L); and ferritin, 8.21 ng/ml (normal range: ≥ 20 ng/ml). Upon physical examination, the patient showed bilateral inguinal lymphadenectasis. A smoking history was reported by himself with a duration of about 20 years (50 cigarettes per day). Nowadays, the patient does not smoke for 20 years. He had a history of drinking alcohol (about 150 g per day) for 40 years.

BM biopsy showed a markedly hypercellular marrow (85%), which was diffusely involved by a small lymphocytic infiltration (80%) (Figure 1A). The infiltration consisted of numerous small lymphocytes and scattered paraimmunoblasts (Figure 1B). The tumor cells were positive for CD20, CD5, and CD23, and were negative for CD3, CD10, TdT, and cyclin D1 (Figures 1C–E). For the aspirate smear, the majority (80%) were small lymphocytes with coarse chromatin. Peripheral blood smear showed scattered, small to medium sized lymphoma cells with irregular nuclear contour. Flow cytometry on BM aspirate demonstrated the presence of kappa restricted B cells, together with positive staining for CD19, CD20, CD22, CD5, and CD23. No CCND1 gene break or P53 gene deletion was identified by FISH, while no mutation was identified in MYD 88 gene after Sanger sequencing.

The patient was then diagnosed with CLL/SLL and microcytic hypochromic anemia simultaneously. Thoracic and abdominal CT showed lymphadenopathy in mediastinum, bilateral neck, axillary area, enterocoelia, retroperitoneal and pelvic cavity, as well as splenomegaly. CT scan indicated partial thickening in the small intestine. Thus, low-stage CLL/SLL was considered, and the patient merely received iron supplement for the treatment. However, the patient still reported persistent needle-like pain in the upper abdomen after treatment. CT enterography showed thickening in the small intestinal wall (Figures 2A–C), together with multiple lymphadenopathies in the abdominal cavity and retroperitoneum. Colonoscopy indicated three polyps with a size of 0.8 cm, 0.7 cm, and 0.4 cm, respectively. Pathological analysis indicated tubular adenoma with low-grade dysplasia. Finally, the patient was discharged with attenuation in the abdominal pain without treatment.

The patient showed recurrence of abdominal pain lasting for 2 weeks after discharge. Abdominal CT showed increase in the diameter and number of lymph nodes. Enteroscopic examination indicated multiple ulcers in the distal jejunum, which were deeper in site with an irregular margin covering them with white pus moss (Figures 2D,E). Based on the biopsy, inflammatory disease was considered by a general pathologist. Then he received laparoscopic exploration and partial enterectomy after endotracheal intubation and combined intravenous-inhalation anesthesia, which indicated multiple ulcers and narrowing in the small intestine. The intestinal wall was narrow and there was a mass at the position that was about 100 cm from the Treitz. Then a part of the intestine including the narrow lesion and the mass (2.5 cm × 1.0 cm × 0.5 cm) was excised. A narrowing site in the intestinal wall was observed at a distance of 2.5 cm from the mass. There was an ulcer in the lesion with a length and a diameter of 2.0 cm and 0.5 cm, respectively. Intestinal edema was noticed between the mass and the narrowed intestinal wall. The villous architecture was distorted in the mass and the whole layer was diffusely infiltrated by neoplastic lymphocytes. The villi were broadly expanded and infiltrated by the tumor cells, presenting prominent epitheliotropism in the peripheral intestine (Figures 3A–C). The cells were moderate in size with a generous rim of the pale cytoplasm. The nuclei were irregular, and most of them showed finely dispersed chromatin and inconspicuous nucleoli. Partial nuclei showed nuclear fold, prominent nucleoli, and dense chromatin. There was no inflammatory background or necrosis. Additionally, small lymphoid cells were seen around the blood vessels in the intestinal wall (Figure 3D).

The tumor cells were positive for CD3, CD7, CD8, CD43, CD56, TIA1, Granzyme B, and Bcl-2. In addition, the tumor cells were negative for CD20, CD79a, CD10, Bcl-6, MUM-1, CD23, cyclin D1, CD2, CD4, CD5, and CD30. The Ki-67 index was in a range of 70%–80% (Figures 3E–L). Tumor cells near the blood vessels were positive for CD20, CD79a, CD5, CD23, CD43, and Bcl-2. The Ki-67 index was in a range of 5%–10% (Figures 3M–P). The results were summarized in Table 1. At the focal edge area of the mass, there was mixed immunophenotype expression. The expression of EBV was negative in the tumor mass and the cells in the peripheral vessels. Finally, the patient was diagnosed with composite lymphoma with MEITL and CLL/SLL simultaneously. The peripheral blood and BM were not invaded by MEILT tumor cells based on the aspiration and the BM biopsy, which was confirmed by flow cytometry to the BM and peripheral blood.

There was clonal rearrangement for the IG detection using specific BIOMED-2 primers based on peripheral blood sample. Sanger sequence confirmed that it was IGHV3-33_06. The homologous degree was 100% compared with the germline sequence. PCR-based TCR assay revealed a small monoclonal T-cell population in a background of oligoclonal T-cells.

For the next-generation sequencing (NGS), B- and T-neoplastic cell content was estimated based on morphology, immunohistochemistry (IHC) and IG/TCR gene rearrangement. Two different paraffin blocks, with about 20% neoplastic cells in the HE sections, were selected for IG and TCR gene rearrangements and targeted sequencing. IHC indicated CLL/SLL harboring IG gene rearrangement, and MEILT harboring TCR gene rearrangement. Enrichment for areas of interest was scraped manually when comparing with the HE staining. High-throughput sequencing analysis was performed using haematopoietic and lymphoid specific panels covering the coding sequence (CDS) of 143 Hematological Disease genes through Illumina NextSeq 550 with a mean sequencing depth of 1000× (SINO-US Diagnostics Lab, Tianjin, China) (Supplementary Table S1). Data were analyzed using the bioinformatics pipeline in-house.

According to the NGS results, the pathogenic hotspot mutations of CLL/SLL and MEILT were STAT5B (c.1924A>C, p.N642H) and DNMT3A (c.2678G>A, p.W893*). Additionally, gene mutation on BCOR (c.1005dupC, p.S336Lfs*45) was only detected in the CLL/SLL area. The likely pathogenic mutations of CLL were SETD2 (c.4688delG, p.G1563Afs*2), NOTCH1 (c.7541_7542del, p.P2514Rfs*4), SF3B1 (c.2902-2A>T), and PTPN11 (c.1492C>T, p.R498W). The TET2 (c.4094G>T, p.G1365V) and ZRSR2 (c.1237A>G, p.K413E) mutations were likely pathogenic for the MEILT. Furthermore, mutations on GATA3 (c.124C>T), PLCG2 (c.1311T>G), and FAT1 (c.7130C>T) were detected in both CLL/SLL and MEITL areas, which were classified as variants of uncertain clinical significance (UVS) (Table 2).

For the follow-up, the patients did not receive additional treatment and passed away on January 6, 2022. This study was performed according to the convention of the Declaration of Helsinki, and the study protocols were approved by the Ethics Committee of Qilu Hospital (Qingdao). Written informed consent was obtained from the patient and his guardians.

Composite lymphoma is defined as coexistence of two distinct types of NHL or a rare combination of HL and NHL in a single organ or tissue (3), comprising 1%–4% of malignant lymphomas (4). Discordant lymphoma is defined as two distant sites involving two different lymphomas. In cases of sequential occurrence of two different lymphomas, it is known as a secondary lymphoma. Our patient was first diagnosed with CLL/SLL, and then was diagnosed with MEITL 1 month afterwards. The abdominal symptoms were presented at the first visit and the two NHLs were presented in the intestinal wall simultaneously. Therefore, the patient was diagnosed with composite lymphoma with CLL/SLL and MEITL.

In 1992, Strickler et al firstly reported 2 cases with nodal peripheral T cell lymphoma (PTCL) concurrent with CLL/SLL (5). Since then, 36 additional cases had been reported, among which 21 cases showed composite lesions involving both PTCL and CLL/SLL in the same biopsy (6). Among these 38 cases, 22 cases (57.9%) were qualified as PTCL or PTCL nos, including 12 with a cytotoxic phenotype, while 12 cases (31.6%) were ALCLs including 6 ALK-positive, 5 ALK-negative, and 1 ALK status unknown. The other 4 cases (10.5%) showed aggressive EBV-negative natural killer (NK)-cell leukemia (n = 1), EBV-positive nasal NK/T-cell lymphoma (n = 1), T-large granular lymphocyte leukemia (n = 1), and nodal lymphoma with a TFH cell phenotype (n = 1) (6). MEITL is a cytotoxic CD8⁺ T cell lymphoma, which accounts for the vast majority of primary intestinal T-cell lymphoma in Asia. Most MEITL cases show jejunal and ileal involvement. The neoplastic cells are featured by medium-sized round nuclei with a rim of pale cytoplasm, which usually infiltrate the intestinal epithelium. Our patient presented abdominal pain, and the medium-sized tumor cells infiltrated the intestinal wall with prominent epitheliotropism. Finally, the patient was confirmed with CLL and MEITL based on the immunophenotype analysis.

The patient was misdiagnosed as Inflammatory disease at first based on biopsy by a general pathologist. After reviewing the biopsy slides, we noticed atypical lymphoid cells in the background of necrosis and inflammatory exudation. Some tissues were crashed and the morphology was indistinguishable. Besides, there was no epitheliotropism in the biopsy. These may lead to misdiagnosis based on the biopsy. Furthermore, attention should be given to the morphology of the tissues, in order to avoid misdiagnosis.

In our patient, STAT5B, DNMT3A, GATA3, PLCG2 and FAT1 mutations were identified in both CLL/SLL and MEITL tumor cells. Activating mutations in STAT5B have been identified in a high proportion of MEITL cases (7-9). In a previous study, Diamantopoulos et al showed that the expression of STAT5B was correlated with the presence of EBV and LMP1, which was negatively correlated with the overall survival of the CLL patients (10). Somatic DNMT3A mutations were rarely identified in CLL and MEITL patients, however, low DNMT3A expression was associated with the pathogenesis of more aggressive diseases (11, 12). Trimech et al reported a small series of patients with composite lymphomas consisting of CLL/SLL and angioimmunoblastic T-cell lymphoma (AITL), in which the AITL comprised prominent clear cells with similar mutations consisting of TET2 or DNMT3A alterations (6). DNMT3A mutation was identified in both CLL/SLL and MEITL tissues in our case. As the expression of VAF was low, DNMT3A might be a manifestation of clonal hematopoiesis of indeterminate potential (CHIP) in case of the patient. In addition, TET2 mutation was found in MEITL area rather than the CLL area, which was not frequent in MEITL patients (12). As the expression of VAF was 8.4%, we think TET2 mutation might be a passenger mutation or a manifestation of CHIP in these patients. Mutational analyses suggested that CLL progression on ibrutinib was closely related to the mutations of BTK and/or PLCG2 genes (13). Monica et al reported that 10.3% of fludarabine refractoriness (FR)-CLL cases showed mutations of FAT1 gene that encoded a cadherin-like protein involving in the negative regulation of Wnt signaling. On this basis, FAT1 mutation may play important roles in the development of a high-risk phenotype (14). In our case, PLCG2 and FAT1 mutations were also detected in MEITL areas, and GATA3 mutations were identified in CLL and MEITL. GATA3, PLCG, and FAT mutations showed similar VAFs in CLL and MEITL tissues, which indicated that the mutations in this patient were germline variants.

Mutations of SETD2, NOTCH1, SF3B1, BCOR, and PTPN11 were identified in the CLL/SLL area of our patients. In a previous study, Parker et al. identified recurrent deletions of the SETD2 locus in 3% (8/261) of CLL patients, and they detected mutations of SETD2 in an additional 3.8% of patients (23/602) based on NGS results (15). Their data highlighted SETD2 aberration as a recurrent, early loss-of-function event in CLL pathobiology linked to aggressive diseases (15). However, the alterations in SETD2 gene were more frequent in MEITL as recorded in 93% of Western European cases (16), and in 70% of cases from North America (17). In contrast, Chen et al. only reported two cases with SETD2 mutation in Chinese MEITL patients (8). In our case, SETD2 mutation was found in the CLL/SLL area, not the MEITL area. Mutations in NOTCH1 and SF3B1 were associated with a poor prognosis in CLL/SLL cases (1, 18). The SF3B1 variant showed a very low allele burden and loss of function, which was not likely to be an oncologically relevant variant in this patient. Previously, Richter’s transformation had been considered to be associated with NOTCH1 mutations (1, 19), and BCOR mutation was detected in 6.25% of CLL patients (20). This was the first report on PTPN11 mutation in CLL and ZRSR2 mutation in MEITL. However, the VAF of PTPN11 and ZRSR2 is very low, and further studies are required to investigate the unknown significance.

The following aspects may help to explain the pathology of composite lymphomas. Firstly, there was an underlying chronic immune dysregulation and T-cell stimulation in CLL patients. Certain factors secreted by CLL cells (e.g. inflammatory cytokines), may chronically stimulate normal lymphocytes, which then could lead to neoplastic transformation (21). In addition, studies on cellular immunity in patients with CLL/SLL reported reduced T-cell function with a paradoxical clonal expansion of CD8⁺ T cells and increased circulating CD8⁺ T-cell counts (22). This might explain the fact that the majority of T-cell lymphomas in CLL/SLL patients may have a cytotoxic phenotype, which expressed CD8 and/or cytotoxic granule proteins, in contrast to T-cell lymphomas in the general population (2, 6, 23-25). Secondly, composite lymphomas may be transformed from a stem cell with the possibility of developing into B-cell or T-cell lineages under different stimulations. Thirdly, the composited malignancies may not be directly related but may represent a shared genetic predisposition or step-by-step oncogenic potential. In cases with simultaneous diagnosis of CLL/SLL and T-cell lymphoma, it is not possible to determine which is primary. Nevertheless, in the present case, no specific treatment regimen was given to the CLL prior to diagnosis of MEILT. Additionally, both CLL and MEITL harbored STAT5B, DNMT3A, GATA3, PLCG2, and FAT1 mutations. All these suggested that these two lymphomas may be derived from the same progenitor cells, while the different gene mutations might indicate that the progenitor cells received different environmental stimulation.

In summary, we report a rare case of composite CLL/SLL and MEITL that highlights the importance of careful histopathologic inspection and immunophenotypic features of hematologic neoplasms. We also highlight the utilization of NGS in screening different gene mutations in CLL and MEITL tissues.