Lymphoplasmacytic lymphoma presenting cold agglutinin syndrome: Clonal expansion of KMT2D and IGHV4-34 mutations after COVID-19

Cold agglutinin disease (CAD) is an autoimmune haemolytic anaemia (AIHA) mediated by monoclonal IgM anti-I autoantibodies without significant lymphoproliferative disease (LPD). Cold agglutinin syndrome (CAS) is associated with various LPDs.^1-3 CAD is categorized as an independent disease according to the fifth edition of the World Health Organization (WHO) classification of haematolymphoid tumours¹ and the International Consensus Classification (ICC) of mature lymphoid neoplasms.⁴ Recurrent somatic mutations in KMT2D and CARD11 have been detected in CAD, whereas mutations in MYD88 L265P and CXCR4 and very rarely KMT2D mutations are detected in patients with lymphoplasmacytic lymphoma (LPL).^{2, 5-8} The cold agglutinins (CAs) encoded by IGH gene segment V4-34 can bind to the I/i carbohydrate antigen on the surface of erythrocytes.^{2, 9} However, the mutational status of IGH gene in LPL with CAS is still unclear. Furthermore, previous reports have indicated an association between CAS and coronavirus disease 2019 (COVID-19).^10-12 The mechanism underlying CA production in COVID-19 remains to be elucidated. Here, we present a crucial case that sheds light on the pathophysiology of LPL with CAS associated with COVID-19.

A 58-year-old Japanese male consulted an ophthalmologist owing to sudden onset of a visual field defect in his right eye. Subsequently, his immunoglobulin IgM was markedly elevated at 9094 mg/dL, while IgG and IgA were within normal limits. Serum immunofixation electrophoresis (IFE) showed the IgM-kappa type. His serum concentration of the kappa chain was 75.2 mg/dL, and that of the lambda chain was 7.2 mg/dL, resulting in a kappa/lambda ratio of 10.44. Urinalysis revealed the presence of kappa type Bence Jones proteins, suggesting IgM monoclonal gammopathy. Laboratory tests revealed a white blood cell count (WBC) of 8700/μL (with neutrophils accounting for 45% and lymphocytes for 38%), a haemoglobin concentration (Hb) of 8.2 g/dL, platelet count of 121 000/μL, total bilirubin of 1.6 mg/dL, indirect bilirubin of 0.8 mg/dL, lactate dehydrogenase (LDH) of 180 U/L and haptoglobin <1.0 mg/dL. Moreover, the patient exhibited a positive result for CA (titre >2048-fold) and C3 in a direct globulin test. All antibodies for infections that cause CAS, such as anti-Mycoplasma pneumoniae antibodies, human immunodeficiency virus (HIV) antigen/antibodies, hepatitis B virus (HBV) surface antigen and hepatitis C virus (HCV) antibodies, were negative. Cryoglobulin was also not detected. Additionally, BM aspirate revealed monotonous proliferation of small- to medium-sized lymphocytes; plasmacytoid lymphocytes accounted for 38%, and plasma cells accounted for 4%, along with erythrocyte hyperplasia with severe haemagglutination (Figures 1A and 2B). Immunostaining of BM biopsy specimens was performed and confirmed the following immunophenotypes: CD20 (+), CD3 (−), CD5 (+), CD10 (−), Bcl-2 (+), Bcl-6 (−), CD138 (+), CD23 (−), Cyclin D1 (−), CD56 (+), kappa chain (+), lambda chain (−), IgM (+) and Ki-67 < 5%. Consistent with the distribution of CD138-positive cells, IgM was positive; kappa chains were positive in both CD20- and CD138-positive cells (Figure 1B–G). Based on these results, the patient was diagnosed with LPL and CAS. Positron emission tomography-computed tomography (PET-CT) identified abnormal accumulations in the vertebral body, iliac bone and spleen, but no definitive nodular lesions were observed (Figure 1H).

Because of high levels of IgM, the patient was started on a single-agent chemotherapy regimen with bendamustine, considering the possibility of IgM flares with rituximab.¹³ After two cycles, the IgM level decreased to below 4000 mg/dL, and the patient was then treated with bendamustine and rituximab. However, the patient developed COVID-19 during treatment and experienced a haemolytic attack. Administration of lemdesivir and sotrovimab led to improvement in his COVID-19 condition without any severe complications. Despite a significant decrease in both IgM levels and LPL cells in the BM, as well as the undetectable expression of kappa chain determined by flow cytometry, levels of CA did not diminish, and the patient experienced recurrent haemolytic attacks after COVID-19 (Figure 2A).

Therefore, whole-exome sequencing (WES) and IGH gene rearrangement analyses were conducted using genomic DNA extracted from BM mononuclear cells at diagnosis and after COVID-19 to explore the molecular diagnosis as well as to assess clonal evolution during COVID-19 infection. The WES results indicated the presence of a KMT2D mutation (c.13885A>C, p.Thr4629Pro), with a variant allele frequency of 9.5% (coverage 57) only in BM after COVID-19 and not in BM at diagnosis (coverage 47) (Figure 2B). Neither the MYD88 L265P nor CXCR4 mutations characteristic of LPL were detected (Figure 2B). Additionally, somatic hypermutation of the IGHV4-34 gene was observed in both BM mononuclear cells at diagnosis and after COVID-19.

Thus, we obtained critical insights into the pathogenesis of CAS using WES and IGH rearrangement analysis for the present case. As previously reported, cases of LPL with CAS have been documented.¹⁴ Almost all LPL patients carry MYD88 L265P and/or CXCR4 mutations, whereas the presence of KMT2D mutations is rare.^{8, 14} In the present case, potentially refractory KMT2D mutant clones might have expand after COVID-19. The emergence of CA after COVID-19 does not necessarily cause significant haemolysis in the clinic. Among patients with AIHA associated with COVID-19, those who died had multiple underlying medical comorbidities, including haemolytic risk factors.¹⁵ In the present case, the appearance of latent clones carrying IGHV4-34 mutations might lead to notable haemolysis (Figure 2C).

In cases of LPL with CAS, MYD88 and CXCR4 mutations are not necessarily present. They may be associated with IGHV4-34 and KMT2D mutations, as in the present case. Clones harbouring these mutations could demonstrate treatment resistance and susceptibility to COVID-19. While next-generation sequencing, including WES, is not yet routinely performed in clinical practice, clinicians should exercise caution when treating LPL with CAS due to the potential aggravation of CAS in the context of COVID-19.

Kiyohito Hayashi performed the experiments, analysed the data and drafted the manuscript. Daisuke Koyama designed the study, performed the experiments, analysed the data and wrote the manuscript. Yuki Sato and Masahiko Fukatsu performed the experiments. Takayuki Ikezoe supervised the research and wrote the manuscript.

The authors have no conflicts of interest to disclose.

The study was performed with the approval of an Institutional review board, and written informed consent was obtained from the patient.