MYC Immunohistochemistry Assessment in High-Grade Lymphomas

Between 10% and 15% of diffuse large cell B-cell lymphomas (DLBCLs) are MYC driven. When MYC alterations are associated with either BCL2 and/or BCL6 rearrangements, they are labeled as double hit (DH; MYC-BCL2 or MYC-BCL6 ) or triple hit (TH; MYC-BCL2-BCL6 ). This distinction, of course, is immediately relevant for current clinical management of these neoplasms because DH and TH DLBCLs respond poorly to the standard chemotherapy regimen which includes rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone. 1 Most MYC -driven lymphomas, including Burkitt lymphoma (BL) and DH, TH, and the unclassifiable aggressive B-cell lymphomas, harbor a balanced translocation involving MYC and the heavy-chain locus: t(8;14)(q24;q32). The latter was first demonstrated in BL 2 40 years ago using karyotyping and is now detected on paraffin-embedded tissue using fluorescence in situ hybridization (FISH), mostly with dual-color break-apart probes (one centromeric probe and one telomeric probe related to the MYC gene). Despite small divergences among laboratories in terms of the percentage of cells with MYC break-apart signals as an analytic cutoff for defining an actionable result, FISH is highly reproducible and specific. Despite being expensive, labor intensive, and time-consuming, it currently represents the gold standard to detect subsets of DH and TH among DLBCLs that morphology alone cannot distinguish. However, the break-apart FISH technique does not detect all MYC -driven B-cell lymphomas. Some uncommon translocations involving far centromeric breaks and some cryptic insertions of the heavy-chain locus into the MYC locus 3 can be missed by current FISH probes. MYC amplification is usually detected by FISH but ideally would require specific probes. 4 Between 2% 5 (if defined as five or more MYC gene copies per nucleus) and 15% 6 (if defined as three or more MYC gene copies per nucleus) of DLBCLs would be associated with MYC amplification. That being said, even only 2% would represent up to 20% of chromosomal anomalies responsible for MYC -driven B-cell lymphoma. FISH could also miss several DLBCLs that have MYC protein upregulation independent of gene alterations: microRNA alterations 7 are relatively frequent and are not detected by FISH. Therefore, independent of the involved molecular anomaly, dysregulated MYC protein overexpression is virtually expected in all MYC -driven B-cell lymphomas, which explains persistent interest in MYC immunohistochemistry (IHC). In 2012, Kluk et al 8 described an IHC method for identifying DLBCLs with increased MYC protein expression. The study contained a small number of cases (n = 77 DLBCLs, with 15 cases with high MYC protein expression), but all MYC translocation-positive cases were identified with a cutoff of 50%. Furthermore, the 50% cutoff showed a significant difference in terms of overall patient survival. In this issue of the American Journal of Clinical Pathology , the same author, together with colleagues from several academic institutions in the Boston area, revisits the IHC technique in the context of current subspecialty-based expert clinical practice. 9 Not surprisingly, increasing the number of observers analyzing an identical data set introduces interobserver variability. The previous two-tier classification based on a 50% cutoff was found not sufficiently robust for clinical practice, and a new three-tier classification was suggested and defined as follows: 30% or less (MYC-IHC low), more than 30% and less than 70% (MYC-IHC indeterminate), and 70% or more (MYC-IHC high). The three-tier system was associated with the highest interobserver agreement. Of interest, level of expertise was not a key factor in interobserver agreement, as the authors found similar interobserver variability (between expert and nonexpert) compared with intraobserver variability (among experts only). As cited by Kluk and colleagues, 9 similar interobserver concordance with MYC IHC has been reported in a recent study from Mahmoud and associates 10 at the University of New Mexico. On the basis of these observations, together with direct comparison with FISH data, Kluk et al 9 concluded that MYC IHC determined as a percentage of positive nuclei cannot be used alone to predict MYC -driven B-cell lymphomas but should be used in conjunction with FISH. To this end, they suggest (in Figure 6) a new algorithm involving both FISH and IHC. An interesting parallel is found in breast pathology, which helps to understand interobserver variability. Percentage of Ki-67–positive nuclei in breast carcinoma is expected to identify, with a reasonable degree of precision, the luminal B breast tumors that are associated with a dismal outcome. A recent international study 11 has attempted to standardize visual assessment of Ki-67 by providing instructions on staining thresholds and a prescribed scoring pattern to determine the percentage of stained tumor cells scored by different breast pathologists. Despite some progress, especially compared with a prior study performed by the same group, 12 the authors showed a significant limitation: irreducible subjectivity biased results when assessors scored faint levels of nuclear staining. Adopting breast pathology good laboratory practice could also improve MYC IHC assessment: IHC biomarkers routinely performed in breast pathology (estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2) are reliable, 13 pending strict control of preanalytical factors that include, for instance, monitoring of cold ischemia, formaldehyde time exposure, and utilization of on-slide controls. In the laboratory, antigen degradation might occur in unstained slides, requiring utilization of freshly cut preparation. 14 So far, this has not been undertaken to improve MYC IHC assessment reliability, although Kluk and colleagues 9 discuss the importance of controlling preanalytic variables in this setting. It takes substantial effort in the medical community to establish an optimal tissue handling practice, but this is the price to pay to validate and implement a class II predictive biomarker. 15 Keeping the same three-tier classification and using strict control of preanalytical variables, it is not impossible that Kluk et al 9 would see an overall stronger and more stable MYC IHC stain with a shift in positive MYC rearrangement cases higher in the histogram presented in Figure 4, which would translate to a more accurate immunohistochemical test. Readership might perceive some push back to MYC IHC utilization, and one might want to build on more promising technology such as molecular classification of MYC -driven B-cell lymphomas based on gene expression profiling (GEP) of fixed biopsy specimens. Unfortunately, according to a preliminary study, 16 GEP is not yet ready to impose itself as a tool to predict MYC -driven B-cell lymphomas: while most BLs were associated with a “high molecular score,” sensitivities to detect MYC rearrangement were 76% and 69% with and without a subset of BL, respectively. Actually, one possible issue related to this GEP is the contamination of samples by nonlymphomatous cells. Again, another parallel can be made with breast pathology: to palliate poor reproducibility of Ki-67 assessment, Genomic Health’s Oncotype DX (Redwood City, CA) recurrence score is a commercial GEP based on 21 genes 17 (heavily weighted on proliferation genes). Some recent reports have highlighted questionable proliferative assessments by this technique. 18 Most notably, in some well-differentiated, low-grade invasive breast ductal carcinomas, high mitotic activity of nontumoral cells found in cellular stroma artificially increases the proliferative status of the associated tumor. MYC IHC assessment combines protein assessment under direct visualization of tumor cells, and pending improvement in IHC, standardization might yet provide more accurate information. MYC IHC will likely remain an important tool in classification of MYC -driven B-cell lymphomas, but a better, more consistent method may ultimately be required for MYC IHC to meet its full potential as a clinical tool in lymphoma management. Manual IHC assessment based on percentage of positive nuclei is commonly practiced in a variety of different pathology fields but has shown limitations, especially in terms of interobserver variability; new approaches or algorithms should be explored involving, for instance, image analysis and other quantitative variable(s). Meanwhile, Kluk et al 9 have cogently summarized the current general status and limitations of MYC IHC assessment based on percentage of positive nuclei and have suggested a workable algorithm that is applicable to current clinical practice.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.