Indolent lymphomas occurring in bone

When considering the management of a rare malignancy, a clinician may adopt the approaches of “lumping” or “splitting.” The lumper applies management principles established for a broad disease category, whereas the splitter looks for potentially unique characteristics of disease subsets. So, for the rare condition of primary indolent non-Hodgkin lymphoma (NHL) of bone, is a specific management approach required, or can it be treated as for any other lymphomas of the same histology and stage?

Differences in behavior have previously been observed not only between nodal and extranodal lymphomas, but also between lymphomas arising in different extranodal sites [1]. One need only look at the specific features of, for example, diffuse large B-cell lymphoma (DLBCL) of the brain and testis to recognize the critical importance of splitting the extranodal aggressive lymphomas according to individual presenting site. The International Extranodal Lymphoma Study Group (IELSG) has played a key role in studying these entities and others [2,3]. Another example of variation by site is the apparent lack of benefit of consolidative radiotherapy (RT) for Waldeyer's ring lymphoma, contrasted with the importance of consolidative RT (but suggested lack of benefit of Mabthera) for DLBCL of bone [4–6].

When considering indolent primary extranodal lymphoma, one first thinks of marginal zone lymphoma (MZL). Extranodal MZL typically presents with localized disease, and the curative potential of local RT has been well described [7]. As is the case with DLBCL, the features of MZL differ according to site of origin, both with respect to known causative factors and regarding the risk of systemic failure after local therapy. MZL arising in the orbit and salivary glands has a higher risk of systemic relapse (often in other MZL sites) than primary MZL arising in the stomach or thyroid [7]. Follicular (FL) and other indolent lymphomas arising in extranodal sites are much less common, and hence less well characterized [8]. Despite this, differences have been described in the natural history of FL, depending on the primary site of origin. For example, it has recently been suggested that localized FL of the duodenum has a particularly indolent course, and can often be managed expectantly [9]. Histological subtype may also influence outcome in localized indolent lymphomas, with at least one prospective study reporting that localized small lymphocytic lymphoma (SLL) had a lower risk of relapse after RT and chemotherapy than FL [10].

So what about indolent primary lymphoma of bone? In this edition of Leukemia and Lymphoma, Govi et al. report a retrospective IELSG study of this clinical entity, which is indeed very rare, representing only 5% of bone lymphomas in their series and previous reports [6,11]. Despite recruiting cases from 32 institutions in 14 countries, reflecting a 20-year experience from 1980 to 2000, the study accrued only 26 cases: 10 with SLL; 10 with FL; and six with lymphoplasmacytic lymphoma. The authors acknowledge the limitations inherent in a retrospective study of this nature, with small numbers, potential selection biases, varied disease stage and treatment and the lack of contemporary central pathology review.

Nonetheless the study offers several valuable messages. First, indolent primary lymphoma of bone is a real entity but rare. Lymphoma occurring in bone is statistically much more likely to demonstrate aggressive histology, and unusual sites of extranodal involvement, such as bone, should raise the index of suspicion for transformation of a pre-existing indolent lymphoma [12]. Thus it is essential that putative cases of primary indolent NHL bone undergo expert pathological review to rule out more aggressive histology. Positron emission tomography (PET) scanning may assist in obtaining representative tissue when transformation is suspected [13]. Second, five of eight patients who received RT for localized disease remain free of disease with a median follow-up for the entire study cohort of 58 months, whereas all three patients treated with chemotherapy alone for localized disease relapsed. This supports the view that, similarly to localized indolent NHL in other sites, indolent lymphoma of bone has the potential for long-term remission, and likely cure, with regimens that include irradiation. Third, the study suggested that the histological subtype was of prognostic significance, consistent with previous studies indicating that localized SLL may have a better outcome than FL [10]. Overall, patients with advanced disease stages had a similar outcome to patients without skeletal involvement, suggesting that advanced stage disease with bone involvement can reasonably be managed according to histology.

The patients accrued in this study were all staged with computed tomography and bone marrow examination, but in contemporary practice, staging of patients with potentially localized indolent lymphoma should also include PET scanning, which often impacts on staging and may be particularly valuable in identifying additional sites of bone involvement [14,15]. A specific issue in bone lymphoma is the staging and management of unifocal versus multifocal bone disease. Mono-ostotic aggressive NHL appears to have a better outcome then poly-ostotic disease, perhaps, at least in part, due to the inability to give consolidative RT to multiple sites of disease [16,17]. The same consideration likely applies to indolent lymphoma of bone. A variety of RT doses and field sizes were utilized in the study cohort, with a median dose of 35 Gy, which is probably higher than is necessary for local control in indolent NHL [8]. In the future an examination of potential molecular features of indolent bone lymphoma would be of interest, although beyond the scope of the present study [18,19]. In conclusion, the data presented in the IELSG report support the “lumpers,” being consistent with the view that indolent lymphoma of bone can be treated according to the principles used in the management of indolent lymphoma arising in nodal and most other extranodal sites. This information should assist clinicians in the management of this rare condition.

Potential conflict of interest

A disclosure form provided by the author is available with the full text of this article at www.informahealthcare.com/lal.