The Many Faces of Follicular Lymphoma

Follicular lymphoma (FL) is the most common type of indolent non-Hodgkin lymphoma in the Western hemisphere, and is regarded as a well-defined disease entity. Its clear morphology, immunophenotype and genetic background make diagnosing the disease relatively simple. But recent developments are changing our perceptions of FL, and could affect the prognosis and management of the condition. Is it time to change our thinking?

Diagnosis might seem simple…

In 80 to 90 percent of FL cases, the affected lymph nodes display destruction of the regular parenchyma by atypical follicular structures formed by components of the normal germinal center (GC) – centroblasts and centrocytes. In contrast to the reactive GC, the neoplastic follicles are devoid of regular zonation, with a dark zone mainly consisting of blasts and a light zone largely composed of centrocytes. They also show a predominant infiltrate of centrocytes, with a small or medium number of interspersed centroblasts. Aside from these typical morphological changes, the fundamental biological difference between neoplastic and reactive follicles is the expression of the BCL2 oncoprotein in the neoplastic follicles. This provides an ideal diagnostic tool for proving the neoplastic nature of atypical GCs, and expression of this anti-apoptotic protein is the biological hallmark of FL. BCL2 expression is downregulated in all reactive conditions, permitting GC cells to die if the B cell receptor structures on their surfaces are not optimally fit for antigen recognition and/or antibody formation. In more than 85 percent of FL cases, the overexpression of BCL2 – not seen under normal physiological conditions – is mediated by formation of the t(14;18)(q32;q21) chromosome translocation. Molecularly speaking, this means that enhancer elements of the B cell receptor heavy chain (IGH) gene promoter in 14q32 are juxtaposed to the BCL2 oncogene in 18q21. Formation of this IGH/BCL2 chimeric gene leads to constitutive overexpression of the BCL2 protein in the GC, prolonging the lifespan of suboptimal cells and permitting the acquisition of more alterations, eventually leading to lymphoma.

It is also interesting to note that, depending somewhat on age, around 50 to 70 percent of healthy adults have been shown to carry t(14;18)-positive “FL-like” B cells in the peripheral blood, although in very small percentages. These circulating IGH/BCL2 positive cells are thought to represent the possible soil of lymphoma development, although t(14;18)-carrying patients have only a 5 percent lifetime risk of developing overt lymphoma.

In our work as pathologists, the earliest morphological alteration we recognize related to circulating FL-like B cells is “FL in situ,” or “in situ follicular neoplasia,” in which individual GCs are partly colonized by BCL2-positive B cells harboring the t(14;18) translocation, but show no signs of truly invasive disease. So at first glance, FL appears to be a lymphoid cancer type that is very uniformly defined by its morphology and immunophenotype, has a unifying genetic background, and evidences a clearly defined precursor condition – making diagnosis both easy and reproducible.

…but is it?

Despite the classical hallmarks of FL, the idea that it is a uniform disease has been severely challenged recently. Clinicians have known for some time that the disease can present with astonishingly diverse clinical courses – some patients survive for decades, while others succumb to the disease within just a few months. At the same time, pathologists have started to recognize a remarkable heterogeneity in FL’s morphological features. A workshop organized jointly by the European Association for Haematopathology (EAHP) and the Society of Hematopathology (SH), “Redefining the spectrum of small B-cell lymphomas in light of current technology” (1), has assembled an extensive collection of FL cases (as well as other indolent B cell lymphomas) characterized by a plethora of unusual features that deviate from the disease’s classical definition. The cases, presented by contributors from all over the world, showed all sorts of peculiarities related to morphology, antigen expression patterns, and genetic features of the tumors (2). FLs with a predominantly or entirely diffuse growth pattern, or with features related to monocytoid and marginal zone differentiation of the tumor cells within a follicular background, were presented. Antigen expression patterns varied widely, with the typical GC markers CD10 and BCL6 shown to be present in only a subset of cells – or not at all – requiring the use of other GC markers to arrive at the correct diagnosis. Crucially, many of these cases were found to lack the prototypic t(14;18) chromosome translocation, and these tumors also frequently exhibited particular – and deviating – morphological and/or immunophenotypic features.

The riddle of pediatric FL

Along with the curious cases presented at the EAHP-SH workshop, another exception to nearly all of the rules are FLs that develop in children or young adults – the so-called “FL of pediatric type.” In contrast to their adult counterparts, these clonal lymphoid tumors usually present with localized disease (clinical stages I or II) and have an excellent prognosis, with most patients successfully treated by simple excision of the affected lymph node(s). The majority of cases are composed of medium-sized and large GC cells (grade 2–3), are BCL2 expression negative, and are also devoid of the classical IGH/BCL2 rearrangement.

It is perhaps pediatric FL that most profoundly challenges our ideas about the uniform classification principles of lymphoid diseases. According to the World Health Organization Blue Book rules, lymphoid diseases should have a common morphological pattern, so that different pathologists can use the same methods to recognize them. They should also have a comparable clinical background, so that they can be recognized by clinicians. But we are becoming increasingly aware that in many, if not all, lymphoid cancers (not to mention other cancer types), there is a growing awareness of heterogeneity that is beginning to dramatically affect our treatment approaches. Although FL presenting with localized disease may be treated – successfully, in a sizeable proportion of cases – by radiotherapy without systemic chemotherapy, mounting evidence suggests that no therapy at all may be needed in distinct FL variants, including FL of pediatric type.

Time for a change?

The pivotal question for pathologists is whether it is justified to regard – as we currently do – a disease with such a tremendous spectrum of morphologies, biological features and clinical presentations, as just one disease. For the time being, the current practice of defining FL as one disease with differing subtypes and variants, possibly characterized by a distinct clinical course, seems to be justified, but this attitude may change. Newer insights into the diversity of lymphoid proliferations have already influenced taxonomic principles. In the forthcoming update of the 4th edition of the WHO classification of tumors of hematopoietic and lymphoid tissues, it has been proposed that the term “FL in situ” indicating a malignant tumor, should be changed to “in situ follicular neoplasia,” taking into account the low progressive potential of the lesion. If adopted, this new definition will add a new dimension to the question, “to treat, or not to treat?” and ensure that clinicians are even more aware of the indolent nature of the condition.

It is of pivotal importance for pathologists to stay familiar with the expanding spectrum of follicular lymphomas. Treatment for FL is likely to involve very meticulous decision-making, depending on the recognition of crucial traits that characterize lymphoproliferations with tremendously different grades – from the very indolent to the very aggressive. At the same time, it will be our task to unravel the molecular basis of these differences in order to better classify these often enigmatic lymphoid proliferations.