Cancer Classification Clarification

Establishing what determines one cancer class from another is a tricky and ever-shifting science. So, what are the most recent updates in terms of diagnostic criteria? Here, we present two research stories that cover the complexities of cancer classification!
 

Twenty-twenty-WHO
 

Until somewhat recently, all classification of myeloid malignancies were based on the fourth edition of the World Health Organization’s Classification of Haematolymphoid Tumours, published in 2017. But with a fifth edition freshly unveiled in 2022 (1), and so many genetic advancements taking place between editions, how will these new guidelines affect diagnostic criteria?

One of the more salient changes between editions is that the blast cut-off between myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) with defining genetic abnormalities (DGA) is for the most part gone. For comparison, the International Consensus Classification (ICC) sets the blast cut-off for AML-DGA to 10 percent, while cases with 10-19 percent blasts without DGA are assigned as a new category: MDS/AML (2).

A study presented at the recent American Society of Hematology (ASH) Annual Meeting in 2022 sought to establish how the new guidelines would affect cases of myelodysplastic syndromes and acute myeloid leukemia (3). A total of 1451 cases related to MDS and AML as per the 2017 classification were examined, and, of those, 746 were diagnosed with against the 2022 criteria. The AML-DGA group remained similar to the WHO 2017 but changed in composition. The largest genetic contributors to this group were AML with KMT2A-r, AML with MECOM-r, as well as two newly defined categories – AML with NUP98-r, and AML with other DGA. 

More broadly, the study found that the 2022 classifications places a greater emphasis on genetic definitions rather than pure morphology. These new guidelines, the authors state, make comprehensive genetic analysis of AML and MDS mandatory. Further to this, much of the classification criteria between WHO 2022 and ICC is similar, yet a small number of patients will be affected by differences in exact phrasing. Ultimately, the authors predict that, in their current state, these classifications could lead to diagnostic confusion among physicians and patients.

A diffused diagNOSis
 

Diffuse large B-cell lymphoma not otherwise specified (DLBCL, NOS) continues to be a diagnosis with a multitude of complex implications for patients – and it is still the most common form of the DLBCL malignancy worldwide. DLBCL can be divided into a number of subgroups based on molecular characteristics, and there are equally as many approaches for classification. 

Although these systems are in some ways similar, none have yet to be implemented on a wide scale due to issues with consistent classification across patients. Another paper that was presented at the recent ASH Annual Meeting set out to establish a more consistent and efficient system working off the foundation laid by pre-existing classification algorithms (4).

The researchers used whole genome sequencing (WGS) data from 900 tumors from DLBCL, Burkitt, follicular, and other mature B-cell neoplasms. These data were then clustered with non-negative matrix factorization, which revealed seven distinct subgroups based on genetic characteristics. The authors note that a number of these are similar to groups defined as existing classification systems – though they vary in their scope and refinement, and, in some cases, exist as a merger of two previously separate groups.

By refining the genetic subgroups of established classification systems, the authors believe they have created an improved method that will allow for greater understanding of the most complex aspects of DLBCL, NOS?