An important subset of diffuse large B-cell lymphoma (DLBCL) and high-grade B-cell lymphoma (HGBCL) have MYC and BCL2 rearrangements Large B-cell lymphoma or high-grade B-cell lymphoma with MYC and BCL2 rearrangements are commonly referred to as "double-hit" lymphoma (DHL), which portends more aggressive clinical behavior and inferior progression-free survival compared to other cases of diffuse large B-cell lymphoma, germinal center B-cell subtype, or other high-grade B-cell lymphoma (1–5). Both the 5^th Edition of the World Health Organization Classification of Haematolymphoid tumors (WHO5) (6) and the International Consensus Classification of Lymphoid Neoplasms (ICC) (7) now classify diffuse large B-cell lymphoma and high-grade B-cell lymphoma with concurrent MYC and BCL6 “double-hit” lymphomas separately due to the unclear prognostic significance of this combination, with some studies not showing distinct biology for these cases (1, 2), but other studies demonstrating an association with a poor outcome (3, 8–11). Fluorescence in situ hybridization (FISH) is a frequently used technique to detect MYC, BCL2, and BCL6 rearrangements. Because diffuse large B-cell lymphoma and other high-grade B-cell lymphoma can look identical morphologically to “double-hit” lymphoma, at a minimum MYC FISH must be obtained in every case (see Figure 1 for an example of a case that has similar morphology to diffuse large B-cell lymphoma, NOS on various slide stain preparations, but ended up having both MYC and BCL2 rearrangements). In one survey of cytogenetics laboratories, the most common test strategy (67%) was upfront testing of MYC, BCL2, and BCL6 FISH on every case while fewer labs (26%) performed BCL2 and BCL6 FISH testing only if MYC is rearranged (12). In the same survey study, 56% of laboratories performed MYC break-apart probe (BAP) in combination with IGH/MYC dual fusion probe while 43% of laboratories performed only MYC BAP. MYC and BCL2 IHC are known to be poor predictors of MYC rearrangements at the gene level and polymorphisms have even been shown to create false negative MYC IHC results (13). Data are less clear regarding MYC and BCL6 rearrangements, but many institutions continue to test for BCL6 rearrangements given some data indicating these cases may have an inferior outcome (3, 8–11).

Fine needle aspiration biopsy (FNAB) is increasingly used for triage of lymphadenopathy and diagnosis of lymphoma. Additionally, cell blocks (CB) and smear slides from FNAB can effectively be used for FISH in various neoplasms (14–17). However, only a few, small single institutional studies have described FISH (such as MYC) in FNAB smears and cell block for lymphoma (18–21) and no comparisons to core biopsy or the gold standard excisional biopsy exist to our knowledge. We have two principal aims in this multi-institutional study of FISH performance in diffuse large B-cell lymphoma and high-grade B-cell lymphoma: 1) compare the success rate of MYC, BCL2, and BCL6 FISH hybridization across FNAB with cell block, core biopsy, and excisional biopsy, 2) compare FNAB to either core biopsy or excisional biopsy for FISH from the same patient.

Six academic medical centers participated in this study. The medical centers were assigned data access groups in REDCap (see section below on REDCap) and they consisted of Massachusetts General Hospital (MGH), Memorial Sloan Kettering Cancer Center (MSKCC), San Francisco Veterans Administration Health Care System (SFVAHCS), Stanford, University of California San Francisco (UCSF), and University of Virginia (UVA).

A retrospective search was conducted of the pathology informatics systems with keywords “FISH” AND “diffuse large B-cell lymphoma” OR “high-grade B-cell lymphoma” over the 10-year period from 1/1/2010 to 12/31/2019. A complementary search was performed of cytogenetics lab data for all MYC FISH studies performed on cytology samples to catch cases missed by the pathology data system and then identify paired surgical samples through the pathology archive. Only specimens that had MYC FISH performed were included (n=222); of these cases, 200 had BCL2 FISH performed (90%) and 186 (84%) BCL6 FISH performed. Exclusion criteria included bone marrow biopsy specimens (due to alternative fixation used in some of these specimens and decalcification that could cause false negatives), and body fluid specimens (due to the numerous pre-analytic variables such as fixative type that might influence FISH performance in cell blocks). The age of the patient was recorded for each biopsy, but for Tables 1 , 2 , the age of the patient was determined as the age at the first biopsy specimen. The WHO4R classification terminology “high-grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangements” was originally used for “double hit” or “triple hit” cases in this patient biopsy cohort with an endpoint in 2019 irrespective of whether the morphology was more in keeping with diffuse large B-cell lymphoma or high-grade B-cell lymphoma (22). In keeping with the WHO5 and ICC, we subsequently distinguished MYC and BCL2 rearranged cases from MYC and BCL6 rearranged cases. This manuscript uses the terms high-grade B-cell lymphoma-MYC/BCL2 or high-grade B-cell lymphoma-MYC/BCL6 to distinguish these two groups regardless of whether the morphology was originally interpreted as diffuse large B-cell lymphoma or high-grade B-cell lymphoma.

All data extracted from pathology reports was entered into REDCap (Research Electronic Data Capture), a secure, encrypted online database. Study data were collected and managed using REDCap electronic data capture tools hosted at Stanford (23, 24). REDCap (Research Electronic Data Capture) is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources.

At Stanford Cytogenetics Laboratory, FISH was performed on formalin-fixed, paraffin-embedded (FFPE) sections; the area of interest was circled on the corresponding H&E stained slide by the ordering pathologist. ZytoLight (ZytoVision Gmbh, Bremerhaven, Germany) break-apart probe sets were used for MYC, BCL2, and BCL6 FISH. The MYC probe set was used with orange probe 5’ to MYC and green probe 3’ to MYC on 8q24.21. BCL2 (18q21) probe set included 3’BCL2 in green and 5’BCL2 in orange. The same probe configuration was used for BCL6 (3q27): 3’BCL6 is green, 5’BCL6 is orange. All FISH results were scored in 100 interphase cells. A rearrangement was reported if 10% or more cells showed a split signal. The results of these FISH tests were compared to existing FISH results on a different specimen when available. Paired specimens were either obtained from the same anatomic site or from contiguous sites e.g. neck lymph node draining the thyroid or CNS lymphoma spreading to the eye.

At the MSKCC, FISH analyses for MYC, BCL2, and BCL6 (Abbott Molecular, Des Plaines, IL) were performed following a standard protocol, as previously described (25). The area of interest was usually circled on the slide, and the cytogenetics lab staff reviewed the slide to confirm a tumor-rich area was tested in all cases. For each probe set, 100 interphase cells were analyzed. A rearrangement was reported if 10% or more cells showed a split signal.

At UCSF, FISH analysis was performed on formalin-fixed paraffin-embedded (FFPE) tissue sections. The area of interest was circled by the ordering pathologist. Abbott Vysis Dual Color Break Apart probe sets (Des Plaines, IL) were used and FISH was set up according to the probe manufacturer’s instructions (https://www.molecular.abbott/int/en/vysis-fish-knowledge-center/fish-on-isolated-nuclei-from-paraffin). FISH signals were imaged and analyzed using MetaSystems software (Medford, MA). For each probe set, 50 interphase cells were analyzed. A MYC, BCL2, or BCL6 rearrangement was reported if 6.0% or more of cells showed split signals.

At SFVAHCS, MYC, BCL2, and BCL6 FISH were performed at Quest Diagnostics in San Juan Capistrano, CA. Because this was an external lab, the area to perform FISH was not specified or circled. FISH was performed using the probes specific for 3q27 (BCL6), 8q24.1 (MYC), 14q32.3 (IGH), and 18q21.1 (BCL2) [Abbott Molecular and SureFISH, Agilent DAKO]. The cutoff values for BCL6 rearrangement, MYC rearrangement, and t(14;18) in the paraffin-embedded tumor tissue are 11%, 15%, and 3%. For each probe set, 100 cells were analyzed.

At MGH, FISH was performed on 5-micron sections of FFPE tissue; an H&E section was reviewed to select regions for hybridization that contained a majority of tumor cells. Break-apart probes (MYC: Vysis LSI MYC Dual Color, Break Apart Rearrangement Probe; BCL2: Leica Kreatech BCL2 Proximal Green [18Q001B495] and BCL2 Distal Red [18Q002B550) probes]; BCL6: Leica Kreatech BCL6 Proximal Green [03Q008B495] and BCL6 Distal Red [03Q007B550] probes) were hybridized and used to calculate the number of cells out of 50 scored containing a rearrangement. A rearrangement was reported if more than 15% of cells showed split signals.

A total of 222 specimens from 208 unique patients were identified with a large/high-grade B-cell lymphoma diagnosis and performance of either MYC, BCL2, or BCL6 FISH. The clinical characteristics of all patients are described in Table 1 . The pathologic characteristics of all specimens are compared across different specimen types and corresponding p values in Table 2 . The final large B-cell lymphoma diagnosis was established after FISH resulted e.g. high-grade B-cell lymphoma with MYC and BCL2 rearrangements or Burkitt lymphoma; the breakdown of different diagnoses can be seen under the header “Final WHO Diagnosis” in Table 2 . A breakdown of the diagnoses in our cohort is shown in Figure 2 .

The breakdown of biopsy specimens by data access group (assigned to academic medical centers as previously described), specimen type collected, and specimen type that had FISH performed is shown by a Sankey diagram in Figure 3 . Despite the diversity and complexity of some specimen types, for instance, FNAB with cell block and core biopsy with varying degrees of imaging guidance, only three specimen types ultimately had FISH performed: cell block from FNAB, core biopsy, and surgical excisional or incisional biopsy. Most specimens with both a cell block from FNAB and core biopsy had FISH performed on the core biopsy with one specimen having FISH performed on the cell block. Most specimens undergoing FISH were lymph nodes (n=129), followed by non-nodal sites such as solid organs (n=37) or soft tissue (n=32), among others (see Table 2 ). Of the lymph nodes, most were cervical (n=37), retroperitoneal (n=21), inguinal (n=18), or axillary (n=15). All the indications for biopsy are shown in Table 2 ; the four most common indications were initial diagnosis (n=76), obtaining additional diagnostic tissue (n=43), evaluating for recurrence (n=36), and ruling out transformation (n=58).

FISH detected a MYC rearrangement in 39 of 217 specimens successfully tested (18%), no rearrangement in 174 specimens, 4 variant MYC FISH results (three with 3’signal loss and one with extra 5’ signaling), and 4 unsuccessful hybridizations ( Figure 4 ). One specimen was not evaluated for MYC FISH but was evaluated for BCL2 and BCL6 FISH; this specimen was included in this study because the patient had a prior biopsy in this cohort that was evaluated for a MYC rearrangement, and the additional biopsy was to obtain more tissue for ancillary testing. FISH detected BCL2 rearrangements in 68 of 198 specimens successfully tested (34%), did not show a rearrangement in 129 specimens, 1 extra signal BCL2 FISH result on an add(3), 2 unsuccessful hybridizations and 22 specimens without any BCL2 FISH performed ( Figure 5 ). BCL2 rearrangements were detected in 20 of 37 MYC rearranged cases (54%), and these “double-hit” lymphomas are described in the next paragraph. FISH detected BCL6 rearrangements in 39 of 183 specimens successfully tested (21%), did not show a rearrangement in 141 specimens, 3 specimens with variant 5’ BCL6 signal loss patterns, 2 unsuccessful hybridizations, and 36 specimens without any BCL6 FISH performed. The hybridization failure rate was low for MYC (4 of 221 or 1.8%), BCL2 (2 of 200 or 1.0%), and BCL6 (2 of 185 or 1.1%) probes. Of the 8 unsuccessful hybridizations, 5 (63%) showed tissue limitations such as crush artifact, fibrosis, necrosis, and paucicellularity. FISH results across different biopsy types are shown in Table 2 .

Concurrent MYC and BCL2 rearrangements were detected in 20 of the 197 specimens (10%) that had FISH successfully performed at both MYC and BCL2 loci; for 19 of these 20 specimens, this result established the diagnosis as high-grade B-cell lymphoma with MYC and BCL2 rearrangements (high-grade B-cell lymphoma-MYC/BCL2). One case was indeterminate due to the morphologic differential diagnosis of high-grade B-cell lymphoma-MYC/BCL2 versus follicular lymphoma with MYC and BCL2 rearrangements, which is not regarded as equivalent to “double-hit” lymphoma according to WHO5/ICC. Of note, 4 additional cases (4 of 197 or 2%) had variant MYC FISH results, including 3 with 3’ signal loss and 1 with extra 5’ signaling. All four of these cases with variant MYC FISH results also had BCL2 rearrangements, raising the possibility of high-grade B-cell lymphoma with MYC and BCL2 rearrangements, without a definitive diagnosis. Of the “double-hit” lymphoma cases, 7 had FISH performed on FNA cell blocks, 10 had FISH on core biopsies, and 2 had FISH on the excisional biopsy. The clinical indications for these biopsy specimens were as follows: 7 for ruling out transformation, 7 for initial diagnosis, 4 for ruling out recurrence, and 1 for additional diagnostic tissue. Of note, one of the 19 “double-hit” lymphoma cases showed rearrangements with all three FISH probes MYC, BCL2, and BCL6, which falls in the same diagnostic category as cases with MYC and BCL2 rearrangements only but has been called “triple-hit” lymphoma in the literature. Three specimens of 182 (1.6%) showed MYC and BCL6 rearrangements, which according to the ICC, is diagnostic of the provisional entity high-grade B-cell lymphoma with MYC and BCL6 rearrangements. The WHO5 classification does not recognize this provisional entity but addresses the need for additional data in this patient group.

Of the 222 specimens with FISH performed, 16 paired cytology and surgical specimens were identified from 8 different patients (each patient had exactly one set of paired specimens). An additional 9 patients with paired cytology and surgical biopsy specimens were identified at one institution (Stanford) that had FISH performed on only one set of the pair; completion FISH was performed at Stanford on a research basis on the specimen missing FISH to enlarge the paired specimen cohort. In total, 34 paired specimens were analyzed from 17 patients ( Table 3 ). The 17 cytology cases included 13 FNAB with cell block and 4 FNAB with cell block and core biopsy (listed as core biopsy in Table 3 , because FISH was performed on the core biopsy). The 17 surgical cases were 10 excisional biopsies, 6 core biopsies, and 1 small bowel resection. Of the 102 possible FISH tests (3 different FISH loci across 34 paired specimens), MYC, BCL2, or BCL6 FISH was performed in 100 instances (98%). Patient 13 only had BCL2 FISH performed in both specimens and thus MYC and BCL6 FISH results were not comparable ( Table 3 ). Out of the 49 comparisons drawn between these 98 paired FISH tests, 2 comparisons (4%) were discrepant: both showed a BCL6 rearrangement in the FNA-cell block specimen but not in the paired excisional biopsy from patient 3 or core biopsy from patient 10 (see Table 3 ). No MYC or BCL2 rearrangements were present in any of the discrepant samples. The date of specimens was closely matched for the discrepant pairs, including differences of 5 days and 10 days between the acquisition of the cytology specimen and the surgical specimen. Both patients had the same lymph node sampled by FNA and excisional biopsy or core biopsy. For patient 10, necrosis was noted in the core biopsy with the negative BCL6 FISH.

MYC, BCL2, and BCL6 FISH have highly successful hybridization rates that are similar across different specimen types in this cohort, including FNAB, core biopsy, and excisional biopsy. High-grade B-cell lymphoma with MYC and BCL2 rearrangements was detected in 9% of all large B-cell lymphoma specimens, including one case with rearrangements at all three loci MYC, BCL2, and BCL6; MYC and BCL6 rearrangements were found in 1.6% of specimens. No significant difference was found across biopsy types. Paired cytology and surgical specimens demonstrated 96% concordance at all three MYC, BCL2, and BCL6 FISH loci. FNAB with cell block is an equally effective alternative to core biopsy and excisional biopsy for assessment of MYC, BCL2, and BCL6 FISH, which is required for identification of the clinically aggressive subset of large B-cell lymphomas that carry both MYC and BCL2 rearrangements.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Stanford University IRB protocol 38992. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

JM: Writing – review & editing, Writing – original draft, Visualization, Supervision, Software, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. UA: Writing – review & editing, Writing – original draft, Methodology, Data curation. CB: Writing – review & editing, Writing – original draft, Methodology, Data curation. SC: Data curation, Methodology, Writing – original draft, Writing – review & editing. SG: Data curation, Project administration, Writing – original draft, Writing – review & editing. RH: Conceptualization, Data curation, Supervision, Writing – original draft, Writing – review & editing. CK: Writing – review & editing, Writing – original draft, Supervision, Conceptualization. OL: Data curation, Supervision, Writing – original draft, Writing – review & editing, Conceptualization. SL: Conceptualization, Data curation, Supervision, Writing – original draft, Writing – review & editing. AL: Writing – review & editing, Writing – original draft, Supervision, Data curation, Conceptualization. JM: Formal analysis, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing. YN: Supervision, Writing – original draft, Writing – review & editing. RR: Writing – original draft, Writing – review & editing. ES: Writing – review & editing, Writing – original draft, Methodology, Investigation. JY: Data curation, Writing – original draft, Writing – review & editing. SZ: Data curation, Writing – original draft, Writing – review & editing. DG: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.