Ethical approval was obtained from the Independent Ethics Committee of Zhejiang Cancer Hospital (Hangzhou, China). A total of 29 patients (1 male and 28 female) newly diagnosed with PBL and treated between April 2006 and May 2013 were retrospectively evaluated. All records were considered valuable if there was available data on patient demographics, pathological diagnoses, tumor details, therapeutic outcomes and follow-ups.

The pretreatment workup included obtaining a complete patient history and conducting a physical examination, liver and renal biochemical analysis, complete blood cell count, bone marrow biopsy, and computed tomography of the chest, abdomen and pelvis. Staging classification was performed according to the Ann Arbor classification (17) and histopathological diagnosis was based on the World Health Organization nomenclature (18).

When data were available, the stage-modified international prognostic index (IPI) score was defined for each patient included in the study. This score was established by Miller et al (19) and gives one point each for age, increased serum lactate dehydrogenase (LDH) and Eastern Cooperative Oncology Group (ECOG) performance status 2 or higher.

Following diagnosis of DLBCL using a core needle or surgery, chemotherapy alone or in combination with radiotherapy was administered. The chemotherapy consisted of between 4 and 6 cycles of treatment with cyclophosphamide-doxorubicin-vincristine-prednisone (CHOP) or a CHOP-like regimen. Chemotherapy was administered with or without central nervous system (CNS) prophylaxis, consisting of intrathecal methotrexate or cytarabine. The radiotherapy consisted of treatment with between 15 and 25 site-directed radiotherapy sessions, of between 1.8 and 2.0 Gy/session (total, 30–46 Gy), in the month following the completion of the chemotherapy program. Rituximab was recommended for patients with primary breast DLBCL. For other PBL histological subtypes, treatment was confirmed by the multidisciplinary lymphoma team of Zhejiang Cancer Hospital. The efficacy of treatment was assessed according to the International Workshop to standardize response criteria for NHLs (20).

Follow-up was performed by the oncologic outpatient clinic, and patients or relatives were contacted by telephone. The final follow-up was in June 2015. SPSS (version 17.0; SPSS, Inc., Chicago, IL, USA) software was used for statistical analysis. Kaplan-Meier estimators were used to calculate the overall survival (OS) and progression-free survival (PFS) rates. OS was measured from the date of diagnosis to the date of death or final follow-up. PFS was defined as the length of time from the date of diagnosis to the date of initial disease progression or death. Survival curves were plotted using the Kaplan-Meier estimator and compared using the log-rank test. Univariate analysis was performed to determine prognostic factors. P<0.05 was considered to indicate a statistically significant difference and all P-values were two-tailed.

A total of 29 patients were analyzed retrospectively. The baseline characteristics are listed in Table I. In total, 28 patients were female (96.6%) and 1 patient was male (3.4%). The median age was 50 years (range, 24–69). None of the patients had a previous history of benign or malignant breast disease, or breast implantation. The most frequent presentation was with a palpable mass (96.6%) and 3.4% presented with palpable axillary lymph nodes. Left breast involvement was similar to right (44.8 vs. 41.4%, respectively) and 4 (13.8%) patients presented with bilateral breast involvement. The median tumor size was 4 cm (range, 1–10 cm). A total of 16 (55.2%) patients presented with stage IE disease and 13 (44.8%) with stage IIE. A total of 2 (6.9%) patients presented with B-symptoms. The majority of patients (72.4%) presented with a low stage-modified IPI score of between 0 and 1. The most frequent histopathological types were as follows: DLBCL, 82.8%; marginal zone lymphoma (MZL), 6.9%; anaplastic large cell lymphoma (ALCL), 6.9%; and mantle cell lymphoma (MCL), 3.4%. Germinal center (GC) or non-germinal center (non-GC) phenotypic information based on immunohistochemistry using the Hans method (21) were available in 14/24 patients with DLBCL: GC, 6 patients; non-GCB, 8 patients; and undefined, 10 patients.

The first-line therapy administered is summarized in Table II. The majority of patients (93.1%) received chemotherapy, of which four patients received CNS prophylaxis consisting of intrathecal methotrexate (n=3), or cytarabine (n=1). The chemotherapeutic treatment regime was supplemented with rituximab in 11 patients. Radiation therapy was administered in 13 (44.8%) patients to give a median total dose of 36 Gy (range, 30–46 Gy). Among the 27 patients treated with chemotherapy: 21 (77.8%) exhibited a complete response; 5 (18.5%) exhibited a partial response; and 1 (3.7%) exhibited disease progression.

The median follow-up time for all patients was 66.8 (range, 25.4–110.0) months. By the final follow-up session, 22 patients were alive without lymphoma and 7 patients had succumbed to PBL. A total of 6 patients succumbed to lymphoma-associated mortality, including 1 patient who developed progressive disease during chemotherapy, and 1 patient succumbed to chemotherapy-associated hepatic failure. Among the 5 patients who relapsed, 4 (80.0%) relapsed within the first two years. One patient who presented with bilateral breast involvement developed left breast relapse following lumpectomy and chemotherapy, 2 patients developed lymphoma of the bone marrow, 1 patient developed relapses of the lung and mediastinal lymph nodes, and 1 patient developed lymphoma of the skin. No patients developed relapses of the CNS. Kaplan-Meier estimator analysis predicted the 1-, 3- and 5-year PFS rates of all patients to be 89.7, 79.3 and 74.6%, respectively (Fig. 1). Kaplan-Meier estimator analysis predicted the 1-, 3- and 5-year OS rates to be 96.6, 79.0 and 74.4%, respectively (Fig. 1).

The patient with MZL, who received a lumpectomy and five cycles of treatment with CHOP, was alive and disease-free by the final follow-up session. Of the 2 patients with ALCL, the patient who received a lumpectomy, five cycles of treatment with CHOP and 36 Gy of radiotherapy (18 sessions/day at 2.0 Gy/session), succumbed to lung and mediastinal lymph node relapse after 26.6 months. The other patient, who received hyperfractionated cyclophosphamide, vincristine, Adriamycin and dexamethasone/1A alternating with high-dose methotrexate and cytarabine/1B was alive and disease-free by the final follow-up. Of the 2 patients with MCL, the patient who received a lumpectomy and six cycles of treatment with R-CHOP [rituximab (375 mg/m²), cyclophosphamide (750 mg/m²), doxorubicin (50 mg/m²) and vincristine (1.4 mg/m², to a maximum of 2 mg), administered intravenously on day 1 and 100 mg oral prednisone on days 1–5] succumbed to a relapse of the bone marrow after 57.9 months. The other patient, who received a lumpectomy alone was alive and disease-free by the final follow-up.

The value of various potential prognostic factors, including age, ECOG performance status at presentation, tumor size, laterality, LDH levels, Ann Arbor stage, adjusted IPI value, surgery, cycles of chemotherapy received (>4), administration of rituximab and administration of radiotherapy, in predicting PFS and OS were evaluated. The impact of the prognostic factors is listed in Table III. The 5-year PFS rates for patients with bilateral and unilateral breast involvement were 50.0 and 78.4%, respectively (P=0.146 bilateral vs. unilateral). The 5-year OS for patients with bilateral and unilateral breast involvement was 50.0 and 78.1%, respectively (P=0.129 bilateral vs. unilateral). No statistically significant difference was observed in PFS and OS rates between the patients treated with rituximab and those without.