Peripheral T-cell lymphoma diagnosis: building a molecular tool

T-cell lymphoma (TCL) has quite a poor probability of survival (around 25-30% of patients after 5 years), which contrasts with the progress that has recently been made in Hodgkin lymphoma and B-cell lymphoma.31 Some recently defined TCL types, such as anaplastic large-cell lymphoma (ALCL), have a better clinical outcome. However, the majority of cases diagnosed with peripheral T-cell lymphoma (PTCL) will eventually die of the disease, and in some specific tumor types, such as intestinal TCL, the prognosis is even more miserable.

Poor survival probability in this context is associated with serious difficulties in lymphoma diagnosis when using routine morphological and immunohistochemistry tools. PTCL classification involves division into multiple subtypes, typically of low frequency and with hazy distinctions (Figure 1). As a consequence, different studies coincide in achieving a very low rate of reproducibility in TCL diagnosis, especially in recognizing ALK-negative ALCL, and distinguishing between PTCL-not otherwise specified (NOS) and PTCL with TFH phenotype or angioimmunoblastic TCL.4

Figure 1.Peripheral T-cell lymphoma classification following the World Health Organization 2017 Lymphoma Classification, here restricted to the tumor types covered in the article. PTCL: Peripheral T-cell Lymphoma. NK: Natural Killer; ATLL: Adult T-cell Leukemia/Lymphoma; NOS: Not otherwise specified; TFH: T-Follicular Helper; ALK: Anaplastic Lymphoma Kinase; TH: T Helper; AITL: Angioimmunoblastic T-cell Lymphoma.

An important feature of this situation is that the relative frequencies of the tumor types are quite low, which makes it difficult to design and develop clinical trials, and this hampers the introduction of new drugs for PTCL therapy.

Nevertheless, these difficulties have inspired some research groups to provide essential information about the molecular basis of TCL pathogenesis, and to identify some attractive and challenging therapeutic targets.95

Drieux and co-workers,10 in a joint project involving French, Belgian and Swiss hospitals, are now addressing the radical proposal that molecular diagnosis may give a more precise and reproducible way of classifying TCL cases. Using a technique applicable to paraffin-embedded tissue, they measure the expression of 20 genes, including 17 markers relevant to T-cell classification, one Epstein-Barr virus-related transcript, and frequently mutated variants of RHOA (G17V) and IDH2 (R172K/T). Selected genes allow the identification of several entities: TFH cells, the normal counterparts of angioimmunoblastic TCL; TH1 and TH2 phenotypes, which reflect the diversity of PTCL-NOS; T-regulatory cells, for distinguishing ATLL; the cytoxic markers, CD30 and ALK, for identifying ALCL; and CD56 and EBER1, to discriminate T/natural killer (NK)-cell lymphomas.

The results validate the solid basis of the currently used PTCL classification scheme, and highlight the similarity between angioimmunoblastic TCL and PTCL-TFH. The findings show a group of cases with simultaneous expression of TFH markers and TH2 (GATA3), and indicate that ALK-negative ALCL is a heterogeneous condition. Cases of PTCL-NOS appear to be extremely heterogeneous, with the cases divided among the different phenotypes. In this study, ALK-negative ALCL had two distinct profiles, with or without expression of cytotoxic genes.

Clinical correlation confirmed the poor prognosis of PTCL (5-year OS=27%), and the better prognosis for ALK and DUSP22 ALCL, but failed to recognize markers that recognize groups with additional clinical variability. In particular, the division into TH1 and TH2 phenotypes in PTCL-NOS was not found to be clinically significant. It is of particular note that a 90% concordance was obtained between the three centers, thus emphasizing one of the stronger points in this approach, i.e. better reproducibility.

Although this is almost certainly not the last word on the process of creating molecular tools for the routine diagnosis of PTCL, it is an important step forward that establishes the viability of the proposal to integrate gene expression and variant recognition, and raises some questions about the PTCL subclasses as they are currently recognized.

Peripheral T-cell lymphoma diagnosis is an area in which molecular diagnosis can play an important role in the near future. Efforts in this field could prove even more relevant, with the provision, not only of diagnostic markers, but also of predictive therapeutic markers, by which the different lymphoma categories may be associated with the identification of useful targets that can be exploited for therapeutic purposes.

Footnotes

• Funding: This work was supported by grants from the Instituto de Salud Carlos III (ISCIII) of the Spanish Ministry of Economy and Competence (MINECO, RTICC ISCIII and CIBERONC) (SAF2013-47416-R, RD06/0020/0107-RD012/0036/0060 and Plan Nacional I+D+I: PI17/2172, PI16/01294 and PIE15/0081), AECC, and the Madrid Autonomous Community.