Clinical Features and Treatment Outcomes of Non-Hodgkin's Lymphoma with Breast Involvement; Multi-Institutional Analysis of 98 Patients in Korea.

The non-Hodgkin's lymphoma (NHL) with breast involvement is an extremely rare extranodal presentation. The aim of this study is to analyze the clinical features and treatment outcomes of NHL with breast involvement, and to investigate whether the arbitrary classification between primary breast lymphoma (PBL) and secondary breast lymphoma (SBL) has any clinical relevance.

We retrospectively analyzed 98 patients newly diagnosed as NHL with breast involvement from 16 hospitals in Korea between January 1994 and June 2009. The eligibility criteria included: (1) histological confirmation by pathologist should be made, (2) documentation of one or both breasts involvement by histology or imaging modalities was needed. The PBL was defined as disease localized to one or both breasts ± regional lymph nodes (ipsilateral axillary, supraclavicular and internal mammary lymph nodes), and SBL defined as disease with systemic lymph nodes and/or other extranodal organ involvement as well as one or both breasts involvement. Mediastinal and cervical lymph nodes were not regarded as regional lymph nodes. Recurrent lymphomas in the breast following prior treatment were not included in this analysis.

The median age at diagnosis was 45 (range, 17-83) years, and median follow-up duration was 39.2 (range, 0.5-186.0) months. The two most common histologic subtypes included were diffuse large B-cell (68 patients, 69.4%) and mucosa-associated lymphoid tissue (8 patients, 8.2%) histology. Other 7 histologic subtypes were identified. Among 98 patients, 89 (91%) were treated at least 1 cycles of systemic chemotherapy, 82 (84%) treated with anthracycline-based regimens, 44 (44.9%) treated with combination of chemotherapy and rituximab. Any surgery or any radiotherapy to the breasts was performed in 27 (27.6%) patients, respectively. According to the definition, PBL and SBL group were 58 (59.2%) and 40 (41.8%) patients, respectively. The estimated 5-year progression-free survival (PFS) and overall survival (OS) was 51.2% ± 6.8 and 61.3% ± 6.0, respectively. Overall response rate (ORR) of 93 patients who were evaluable was 91.4% (CR, 76.3%; PR, 15.1%). Compared the baseline characteristics of PBL with those of SBL, PBL group showed more favorable clinical factors as 0 or 1 of Eastern Cooperative Oncology Group performance status (p<0.001), normal LDH level (p=0.003), absence of B symptom (p=0.001) and low or low-intermediate international prognostic index (p<0.001). 14 (24%) of PBL group were treated with abbreviated course (£4 cycles) of systemic chemotherapy and local therapy, 11 (28%) of SBL group were treated 4 or less than 4 cycles of systemic chemotherapy because of mainly disease progression or early death. ORR was significantly higher in PBL group (96.6% vs 72.5%, p=0.005), and estimated 5-year PFS (63.6% ± 8.7 vs 35.2% ± 8.9, p<0.001) and OS (71.8% ± 7.2 vs 45.3% ± 9.9, p=0.004) was also significantly longer in PBL group compared with SBL group. Eight (8.2%) patients had central nervous system (CNS) relapse or progression in the course of disease, and PBL group had a significantly higher rates of CNS relapse or progression compared with SBL group (6 [10.3%] vs 2 [5%], p=0.046). In multivariate analysis for PFS, 4 or less than 4 cycles of systemic chemotherapy regardless of any local treatments (hazard ratio [HR], 5.14; 95% confidence interval [CI], 2.18-12.12) and more than 2 of extranodal organ involvement (HR, 10.64; 95% CI, 4.75-23.83) were independent prognostic factors for shorter PFS. And, for OS, 4 or less than 4 cycles of systemic chemotherapy regardless of any local treatments (HR, 4.03; 95% CI, 1.90-8.54) was the only independent prognostic factor for shorter OS.

Although criteria for PBL and SBL, we traditionally used, did not consider tumor biology and an arbitrary definition, we confirmed that the patients with NHL involved breasts could be classified into two different groups; traditional PBL and SBL according to the clinical characteristics, treatment outcomes and patterns of failure in this analysis. Continuation (>4 cycles) of active systemic chemotherapy was the only prognostic factor for OS, regardless of any local treatment. Thus, standard systemic chemotherapy should be the mainstay of treatment for NHL involved breasts. And, newer treatment strategy adapted on the poor prognosis should also be warranted.

No relevant conflicts of interest to declare.