Progressive transformation of germinal centers and Hodgkin lymphoma: more insights but maybe more confusion?

Progressive transformation of germinal centers (PTGC) has been recognized as a form of reactive lymphoid hyperplasia with a predominant follicular pattern, of uncertain nature, for almost 50 years [1–3]. Histopathology shows characteristic enlarged follicles with germinal centers, broken up by an apparent in-growth of mantle type B cells lacking the classical lymphocytic and histiocytic (L&H) cells characteristic of nodular lymphocytic predominant Hodgkin lymphoma (NLPHL) [2,3]. The absence of significant numbers of classical Hodgkin Reed–Sternberg cells, and the presence of a typical immunophenotype that includes strong uniform expression of CD20 and CD79a with positive staining for OCT2, BOB1, heavy and light chains and the absence of CD30 expression, confirm the diagnosis of PTGC [4].

In adults, PTGC occurs mostly in young males and is characterized by painless lymphadenopathy, most commonly affecting cervical, inguinal and axillary nodes, with other nodes or extranodal sites being involved infrequently [5]. Patients typically undergo an excisional lymph node biopsy to establish the diagnosis, and in about 25% of adults diagnosed with PTGC, recurrent lymphadenopathy develops, requiring further biopsies to exclude lymphoma.

There is an association between PTGC and lymphoma, most frequently with NLPHL, and less commonly with low-grade follicular lymphoma or other Hodgkin lymphoma (HL) subtypes. The etiologic relationship between PTGC and these lymphomas still remains unclear, almost certainly partly due to diagnostic uncertainty, particularly in some of the early series. Although the immunohistochemical distinction between these disorders is now well described [3], there is considerable morphologic overlap such that accurate diagnosis may have been problematic in cases dating back to the 1970s and 1980s.

Recent series suggest that about 35% of adult patients with PTGC will also have a history of NLPHL which may precede, follow or coincide with the diagnosis of PTGC. The occurrence of other HL types or follicular lymphoma is less common, representing about 5% of all PTGC. For this reason, once a diagnosis of PTGC is made in an adult, repeated lymph node biopsies are performed because of concern for the development of lymphoma. Although these conditions are closely linked, there is no evidence to suggest that PTGC is a “premalignant” condition. In this issue of Leukemia and Lymphoma, Picardi et al. approach the association between HL and PTGC from a different perspective, investigating the incidence of PTGC in patients treated for HL at their institution [6]. In earlier series, the incidence of this association varied between 1 and 15%, probably reflecting diagnostic inconsistency over time, with the strongest association being with NLPHL. In a 9-year period from 2000 to 2008, the authors treated 350 patients for newly diagnosed HL, and report that 48 (14%) of them subsequently developed PTGC. In the group with NLPHL the frequency of subsequent PTGC was high at 39%. In a historical control group of patients treated for HL in the 1990s, the authors report an identical incidence of subsequent PTGC.

The novel aspect of this study is the use of rituximab in the more recent cohort. Once PTGC was diagnosed, following treatment for HL, they were placed on a standard induction course of rituximab followed by 2 years of maintenance rituximab. When this cohort was compared with the historical control group, who had not received rituximab at the time of development of PTGC, there was a markedly lower rate of lymphoma recurrence in the control group. In all, 13 of 48 “control” patients developed a subsequent lymphoma (most commonly NLPHL), compared with only 1/48 among those receiving rituximab. It is interesting that all lymphoma relapses occurred at sites of previous PTGC involvement.

This report challenges current concepts of PTGC and its management. The high incidence of PTGC after successful therapy of HL is striking, but inevitably raises concerns about diagnostic consistency. There is now a substantial body of literature from single centers, and from multi-center clinical trials conducted worldwide in HL over the last 20 years. Thousands of patients have been entered, with mature follow-up, but PTGC has rarely been reported as an event in the follow-up of these studies.

The observed apparent association between HL and PTGC, particularly the occurrence of HL in lymph node sites identical to those involved with PTGC, has two potential interpretations: either there is diagnostic misinterpretation confusing the issue or PTGC represents a truly premalignant condition – a concept which challenges the current understanding of this entity.

Finally, the authors suggest that the use of rituximab “prophylactically” in this setting can apparently prevent progression of PTGC to lymphoma. This is a provocative conclusion but one which the authors justify on the basis of the apparent lack of toxicity associated with induction therapy and maintenance with rituximab in their experience.

The results of this study are interesting, but physicians taking care of patients with HL need to be circumspect about their conclusions. First, the stated incidence of PTGC, the sequential emergence of this entity and the occurrence of relapsed HL at the same nodal site raise questions about the diagnostic criteria. Furthermore, the use of prolonged therapy with rituximab is controversial because of the probable development of immune impairment including hypogammaglobulinemia and late infectious complications [7].

Finally, a prospective validation of this approach would require a large cohort of patients with HL treated within the context of a prospective study, followed uniformly after achievement of remission, biopsied systematically at relapse with central pathology review and then treated in a prospective, randomized fashion with rituximab versus observation only. The likelihood that such a study will be conducted is low, but until this is done, we feel strongly that the use of rituximab in this setting should be regarded as experimental, and suggest that a “watch and wait” approach should remain the standard of care for patients with PTGC with a high level of awareness of the association with lymphoma.

Potential conflict of interest:

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.