Radiotherapy Alone for Primary Cutaneous B-cell Lymphoma

Abstract & Commentary

Synopsis: There are no well-defined guidelines for the treatment of primary cutaneous B-cell lymphoma (PCBCL). Since it is an uncommon diagnosis, it is unlikely that answers will come from randomized trials. This retrospective study from 2 German centers evaluated outcomes from a cohort of 35 patients and concluded that radiotherapy to all visible lesions is the treatment of choice.

Source: Eich HT, et al. Int J Radiat Oncol Biol Phys. 2003;55:899-906.

Skin is the second most common extranodal location for B-cell lymphoma, after the gut. Primary cutaneous B-cell lymphoma accounts for approximately 5% of all cutaneous lymphomas in the United States. Low-grade subtypes are the most prevalent. According to the 1997 EORTC classification, these would be the follicle center cell (FCCL) and immunocytoma subgroups. Large B-cell lymphoma of the leg (LBCLL) is an intermediate subtype, and plasmacytoma and intravascular B-cell lesions are provisional subtypes.1 PCBCL can present in a variety of ways, and it may take years before a definitive diagnosis is established. Lesions can appear as nodules, papules, or plaques that are usually reddish to livid in color. Eich and colleagues combined data from the Universities of Cologne and Muenster in order to retrospectively evaluate results of therapy.

From 1984 to 2001, 35 patients were treated with primary radiotherapy. All patients with a diagnosis of PCBCL were referred without selection. Median age was 61 years (range, 27-86 years). Sixty-percent of patients had FCCL (n = 21), 20% had immunocytoma (n = 7), 11% had LBCLL (n = 4), and 9% had a provisional variety (n = 3). Thirty-seven percent of patients presented with lesions on their head and neck (n = 13), 23% each on the trunk and lower extremities (n = 8 each), and 17% with PCBCL on their arms (n = 6). Most patients (n = 18, 51%) presented with a single tumor, but 35% (n = 10) had multiple tumors in 1 anatomic region, and 20% (n = 7) had lesions in non-contiguous sites. Lesions varied from 0.5-15 cm in greatest diameter.

The majority of patients received RT alone (n = 29, 83%). The remainder underwent surgery first. All except 2 patients were treated with 5-12 MeV electrons. Two patients were treated with photons, and 1 received mixed photon/electron therapy. Margins on the tumors were > 2 cm for lesions on the head/neck, and 5 cm for lesions on the trunks/extremities. Median dose was 45 Gy (range, 16-54 Gy). Median dose per fraction was 1.8 Gy (range, 1.8-3 Gy). Bolus material was used to bring the dose to the surface of the skin where needed. Electron treatments were prescribed to the 95% isodose line.

All 35 patients, except for 1 whose lesion involved an entire leg and who died after 16 Gy of pneumonia, achieved a complete response to therapy. Treatment was well tolerated in all patients. Median follow-up was 52 months. Median overall survival was 115 months for the entire group, and 5-year actuarial overall survival was 75%. Mean disease-free survival was reported as 77 months, and 5-year actuarial DFS was 50%. At the end of the study period, 27 patients were alive, 18 without disease, and 9 with disease. Three died of disease, and 8 died of other causes. There were 11 cutaneous relapses (31%) after a median of 11 months (range, 3-24 months), 3 of which were in or at the margin of the RT field. Of these, 2 were salvaged with either additional RT or with chemotherapy. Eight patients recurred in skin outside the primary field, and 5 were salvaged with RT, while 3 exhibited progressive disease on chemotherapy. There were no extracutaneous recurrences at first relapse.

Multivariate analysis evaluated age, gender, primary site, histologic subtype, and number of lesions for prognostic significance. Two or more noncontiguous anatomic sites of disease and LBCLL were noted to be independent unfavorable factors.

Eich et al concluded that a total dose of > 40 Gy with > 2-3 cm margins encompassing all visible lesions is the treatment of choice in all patients except PLBCLL, where chemotherapy is advisable. Multifocal skin disease is not an indication for chemotherapy except in the latter subtype.

Comment by Edward J. Kaplan, MD

Bekkenk and associates published their experience with a variety of approaches for 29 patients with PCBCL from the Dutch Cutaneous Lymphoma Group, and the conclusions were exactly the same as Eichs et al.2 Their median follow-up was 50 months. In this study, 55% of patients had FCCL and 28% had the immunocytoma subtype, roughly the same proportions as in the German study.

In contrast to the Eich and Bekkenk studies, Sarris et al from MDACC came to a different conclusion.3 Their retrospective analysis of 46 patients with primary cutaneous lymphoma treated from 1971-1993 contained a preponderance of diffuse large-cell lymphomas (n = 19). Forty-two of the 46 patients had aggressive lymphomas according to the REAL classification. Ten patients were treated with RT alone, and 33 received doxorubicin as part of their therapy. Two patients were not treated at all. The MDACC patients achieved a 90% complete response rate with RT, 100% with chemoradiation, and 88% with chemotherapy alone. At a median follow-up of 140 months, actuarial 12-year disease-free survival was 61%. Seventy-one percent of the diffuse large-cell lymphoma patients had no evidence of disease (NED) at 12 years compared with none who received RT alone (P = 0.0003). FCCL patients enjoyed an excellent 12-year NED rate following chemoradiation. Eich et al concluded that RT alone was not curative for cutaneous diffuse large B-cell lymphoma, and that further work needs to be done in order to define the optimal therapy for FCCL.

At first glance, Eich and the MDACC investigators’ conclusions appear to be at odds with those of the 2 European studies. However, the differences can largely be explained by the patient mix. While the European groups were mainly composed of patients with indolent disease, the American study was heavily weighted toward aggressive lesions. All 3 studies were small. All 3 studies are probably justified in their conclusions. The German and Dutch researchers agreed that patients with large B-cell lymphoma of the leg were not candidates for RT alone, and certainly the MDACC researchers would agree. Whether chemotherapy is necessary for patients with FCCL is not known, but based on the European experience, it may not be required. It will be interesting to watch as more data accrue for patients with cutaneous B-cell lymphomas, but for now there seems to be general agreement that patients with high-grade lesions need chemotherapy integrated into their treatment program, while patients with low-grade disease may do well with RT alone, even in the setting of multiple lesions.

Dr. Kaplan is Acting Chairman, Department of Radiation Oncology, Cleveland Clinic Florida, Ft. Lauderdale, FL and Medical Director, Boca Raton Radiation Therapy Regional Center, Deerfield Beach, FL.


1. Willemze R, et al. Blood. 1997;90:354-371.

2. Bekkenk MW, et al. J Clin Oncol. 1999;17:2471-2478.

3. Sarris AH, et al. J Clin Oncol. 2001;19:398-405.