Lymphoid malignancies in patients with Shwachman-Diamond syndrome Free

TO THE EDITOR:

Many of the inherited bone marrow failure syndromes are associated with myeloid malignancy predisposition, and a subset carry an increased risk of both myeloid and lymphoid malignancies. Shwachman-Diamond syndrome (SDS) is a bone marrow failure syndrome with variable multisystem manifestations.^1-10 Most cases of SDS (>90%) result from biallelic germ line mutations in SBDS.¹¹ Although an increased risk of myeloid malignancy in SDS is known,^2-6,12 the risk of lymphoid malignancy in SDS has not been previously recognized. Through the North American Shwachman-Diamond Syndrome Registry (SDSR), we identified patients with SDS diagnosed with lymphoma. We describe the clinical features, outcomes, and molecular characteristics of lymphoma in SDS below. Additionally, we report the incidence rates of lymphoma, myelodysplastic syndrome (MDS), and acute myeloid leukemia (AML) in 217 patients with biallelic SBDS mutations in the SDSR.

The North American SDSR is a retrospective/prospective cohort study that has been enrolling patients since 2008. Informed consent was obtained in accordance with the registry protocol approved by Institutional Review Boards at Boston Children’s Hospital and Cincinnati Children’s Hospital Medical Center. All patients with biallelic SBDS mutations enrolled on the SDSR were included. Medical records and pathology reports were reviewed to confirm malignancy diagnosis. Age- and sex-specific incidence rates from the Surveillance, Epidemiology, and End Results Program (2016-2020)¹³ were multiplied by corresponding person-years of observation to obtain expected numbers of events. Person-years of observation started at date of birth and continued until diagnosis of malignancy, death, or last day of follow-up. The crude rate per 100 000 (100 000 × observed number of malignancies/person-years of follow-up), the standardized incidence ratio (SIR), and the exact 95% confidence interval for the SIR were calculated. Using the OncoPanel cancer genomic assay,¹⁴ somatic mutation and copy number variant analyses were conducted on tumor DNA extracted from available formalin-fixed, paraffin-embedded lymphoma tissue samples. Using the Rapid Heme Panel assay,¹⁵ targeted next-generation sequencing was performed on bone marrow aspirates.

Of a total of 217 patients with SDS with biallelic SBDS mutations in the SDSR, we identified 4 patients with lymphoma. The demographics, clinical features, and outcomes of these patients are described in Table 1.

We conducted molecular analysis on tumor tissue, which was only available for case 2. This patient initially presented with a κ-restricted diffuse large B-cell lymphoma (DLBCL), germinal center type, of the duodenum at 24 years of age. Bone marrow examination at diagnosis was negative for malignancy but demonstrated an abnormal karyotype: 46,XY,t(11;16)(q23;q24),del(20)(q11.2q13.3) [9]/46,XY[11]. He achieved remission after 5 cycles of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone therapy. Three years later, he presented with distal ileum DLBCL, now λ restricted, concurrently diagnosed with AML with complex karyotype and 5q deletion on microarray. Somatic mutation analysis of the AML marrow identified multiple TP53 clones. He achieved complete remission of the second DLBCL and AML after completing therapy outlined in Table 1 followed by hematopoietic stem cell transplant (HSCT).

Somatic mutation analysis of lymphoma tissue from the initial diagnosis revealed an EZH2 variant at a known mutational hotspot described in germinal-center origin DLBCL,¹⁶ as well as 2 TP53 variants at moderate and near-identical variant allele frequency (VAF) suggestive of co-occurring biallelic alterations. One of these TP53 variants (p.R175H) was also found at low-level frequency in the bone marrow at the time of AML diagnosis 3 years later. Somatic mutation analysis conducted on lymphoma tissue from the second diagnosis demonstrated a new TP53 variant (p.R280G) at high VAF occurring concurrently with a deletion of chromosome 17p, consistent with a biallelic TP53-mutated clone. This same TP53 p.R280G variant was also found at low VAF in the AML bone marrow in the absence of lymphomatous involvement. The TP53 and EZH2 variants identified at initial diagnosis of lymphoma were not detected in the lymphoma specimen from his second diagnosis, and the copy number variant profile differed considerably between the 2 lymphoma specimens, indicating these lymphomas were clonally distinct (Table 2).

Patients with SDS are known to develop somatic TP53 mutations, giving hematopoietic cells a selective, albeit maladaptive, advantage in the setting of ribosomal defect constraints on cell fitness.^4,17,18 Biallelic TP53 inactivation, which can occur by several mechanisms, is the predominant molecular pattern associated with myeloid malignancies in SDS.¹⁷ However, this same mechanistic driver toward malignancy has not been previously described in lymphoid cells in SDS. Interestingly, the κ vs λ clones, different TP53 mutation patterns, and distinct copy number profiles of the 2 lymphoma diagnoses in case 2 indicate the development of a second primary TP53-mutated lymphoma rather than relapse. Additionally, the presence of the p.R175H and p.R280G TP53 variants in both myeloid and lymphoid lineages suggests these variants arose in an early multipotent hematopoietic stem or progenitor cell (Figure 1), although the emergence of independent myeloid and lymphoid clones cannot be ruled out. This highlights the need for further investigation of lineage-specific clonal pathways in genetic cancer predisposition conditions.

A total of 217 patients with biallelic SBDS mutations were included in the analysis of malignancy risk. The median age was 12.8 years (range, 0.3-52.8 years) in our cohort; 62.7% of the patients were male (136/217). We observed 4, 13, and 14 patients with lymphoma, MDS, and AML, respectively. The crude rates per 100 000 person-years were 120 for lymphoma, 396 for MDS, and 420 for AML. The observed risk of lymphoma in patients with SDS was 38 times higher than expected in the general population. The observed rates of MDS and AML were 5409 and 469 times higher, respectively, than expected in the general population. A summary of the observed and expected number of cases, crude rates, and SIRs (observed vs expected cases) for lymphoma, MDS, and AML is presented in Table 3.

Beyond these 4 patients with lymphoma from the SDSR, 2 of whom have been previously described,^2,19 2 additional cases of non-Hodgkin lymphoma in patients with SDS have been reported in the literature.^20,21 Although MDS and AML are the predominant hematologic malignancies in SDS, our analysis demonstrates the incidence of lymphoma among patients with SDS is also increased relative to the general population.

One limitation of this study is potential selection bias. If patients with malignancy are more likely to enroll in the registry than clinically asymptomatic patients, the calculated SIRs would be overestimated. However, SDS is a multiorgan system disease and malignancy is not the sole factor affecting interest in enrollment. Conversely, our results may underestimate true malignancy incidence if underlying SDS is unrecognized in patients diagnosed with a hematologic malignancy. This latter possibility is supported by a study that found that 4% of young adults who underwent HSCT for MDS had biallelic SBDS germ line mutations.²² 

Although all 4 patients with lymphoma from the SDSR achieved complete remission, increased toxicity, including significant infectious and end-organ complications, was observed with standard therapies. Notably, case 2 developed AML 3 years after initial treatment for lymphoma, which highlights the theoretical risk of cytotoxic lymphoma therapy further increasing the already high lifetime risk of myeloid malignancy development in SDS. Nongenotoxic therapies and HSCT following cancer-directed therapy are areas for future investigation to inform clinical recommendations.

In conclusion, SDS should be added to the list of marrow failure conditions predisposing to both myeloid and lymphoid malignancies. To our knowledge, we identified, for the first time, the presence of TP53 variants in lymphoid cells of patients with SDS. The biological and clinical impact of TP53 variant cell of origin as well as the clonal evolution of lymphoma in SDS warrant further exploration. This study also highlights the importance of attention to the high risk of subsequent myeloid malignancy in SDS and the potential for increased toxicity with standard therapies when developing treatment plans for lymphoma in these patients.