Breast Non-Hodgkin’s lymphoma is a highly heterogeneous disease, with DLBCL being the most common pathological type. Related studies have shown that the 5-year OS of breast NHL ranges from 50–75%. In our study, the 5-year OS of patients was 80%, slightly higher than previous literature reports[2, 3]. According to the diagnostic criteria developed by Wiseman et al. in 1972, B-DLBCL can be divided into two categories: PB-DLBCL and SB-DLBCL. PB-DLBCL is defined as only involving the breast, including or excluding the ipsilateral axillary and supraclavicular lymph nodes, while the rest are defined as SB-DLBCL[9]. Based on this standard, the Ann Arbor staging of PB-DLBCL is in stages I-II, with primary lesions mostly located in the breast, belonging to a type of disease with limited involvement in the region; On the contrary, the Ann Arbor staging of SB-DLBCL is in stages III-IV, with primary lesions mostly located in other parts and involving breast tissue during the disease development process. It belongs to the type of disease that involves a wide range of areas and spreads throughout the body; In theory, PB-DLBCL has a better prognosis than SB-DLBCL. In our study, the PFS and OS of PB-DLBCL were also superior to SB-DLBCL, and the findings were consistent with previous reports in the literature [10]. Additionally, DLBCL has a special type of disease, which not only affects the breast and its regional lymph nodes, but also spreads to the blood. It belongs to phase IV PB-DLBCL, and its prognosis is between PB-DLBCL and SB-DLBCL. Yao Sun et al. conducted a meta-analysis and found that classifying B-DLBCL into in situ (PB-DLBCL Ⅰ E - Ⅱ E) and disseminated (PBD-LBCL Ⅳ E and SBDLBCL) was more effective in reflecting the therapeutic effects of the disease[11].
B-DLBCL is insidious and often presents as a painless breast mass. It needs to be differentiated from benign tumor, breast cancer and metastatic tumor of the breast. Therefore, many B-DLBCL patients will be admitted to hospital with the diagnosis of breast cancer. Over the past 20 years, surgery has been one of the main treatment methods for B-DLBCL, but there is still much controversy over whether surgery can be used as the standard treatment for B-DLBCL[2]. Some studies suggest that B-DLBCL is a systemic disease that should be primarily treated with chemotherapy, while surgery as local treatment cannot effectively reduce the overall tumor burden, improve patient prognosis, and prolong survival[12, 13]. For example, by analyzing 465 patients with B-DLBCL reported in the literature, Jennings, W.C et al found that 156 patients who underwent mastectomy did not achieve a higher OS compared to patients without surgery, furthermore, the analysis yielded that the procedure was associated with a trend toward worse OS (P = 0.055)[12]. Similarly, Ryan, G et al through a retrospective analysis of 204 eligible patients with breast lymphoma, showed that there was no benefit from mastectomy, as opposed to biopsy or lumpectomy only. Importantly, radical breast cancer treatment was even associated with an increased risk of death, HR = 2.4 (95% CI:1.2–4.8; P = 0.03)[13]. Uesato, M et al mentioned in a case study that of 380 patients with primary breast lymphoma documented in Japan, 333 patients (87.6%) underwent surgery and found that tumor removal was necessary, but further axillary lymph node dissection (ALND) was unnecessary and the prognosis was better despite enlargement or metastasis at the axillary lymph nodes[14]. However, Mustafa Ali, M.K. et al found that mastectomy with or without regional lymph node dissection was effective in reducing tumor load and improving PFS and OS through a meta-analysis[15]. The reasons for these differences may be due to the relatively small number of patients included in the study and the different surgical options. In summary, whether surgery can be used as a standard treatment for breast lymphoma remains highly controversial. Many studies have concluded that breast lymphoma as a systemic disease should be treated systemically, such as through chemotherapy. Surgery as a local treatment is not effective in reducing tumor load; however, many findings show the positive aspects of surgery. In our study, 13 patients underwent surgery, of which only 1 patient underwent mastectomy combined with ALND, a ratio consistent with that reported in the above study. Although the results showed that the PFS and OS were higher in the operated patients than in the non-operated patients, there was no statistically significant difference (P > 0.05), probably due to the small sample size in this trial, which needs to be further validated by expanding the sample size.
The current standard chemotherapy regimen for B-DLBCL is CHOP (doxorubicin, cyclophosphamide, vincristine, prednisone) regimen, which is consistent with the chemotherapy regimen used in the patients included in this trial. Rituximab is a chimeric monoclonal antibody against CD20 B-cell antigen that is therapeutic for diffuse large B-cell lymphoma. The study showed that standard chemotherapy regimens combined with rituximab improved the prognosis of B-DLBCL, especially for non-GCB B-DLBCL, eliminating the difference in prognosis between non-GCB and GCB types[16, 17]. According to Coiffier, B et al conducted a randomized trial comparing the efficacy of CHOP chemotherapy plus rituximab with CHOP regimens alone in the treatment of elderly diffuse large B-cell lymphoma in groups, and the results of the study suggested that rituximab significantly improved complete remission rates, prolonged event-free survival and overall survival in elderly patients with diffuse large B-cell lymphoma with no significant increase in clinical toxicity[16]. Consistently, Pfreundschuh, M et al analyzed the efficacy of rituximab in young DLBCL patients by a similar approach and came up with the same results[17]. It is interesting to note that Bouwstra, R et al showed the opposite result with their study that although the addition of rituximab(R) to CHOP chemotherapy regimens improved the prognosis of patients with DLBCL, 40% of patients still failed treatment due to drug resistance[18]. A similar conclusion was reached by Aviles, A N et al, non-GCB genotypes are more common in PB-DLBCL, but the addition of rituximab did not improve the prognosis of primary breast lymphomas with non-GCB phenotypes[19]. The results of this trial were similar to those of Aviles, A N et al, the use of rituximab did not result in improved PFS and OS in patients.
CNS recurrence or metastasis is the most common cause of treatment failure in DLBCL, and extra-nodal DLBCL is more likely to develop CNS metastasis. Methotrexate (MTX) is the most widely used treatment regimen to prevent central nervous system (CNS) metastasis or recurrence in patients with DLBCL. However, Bobillo, S et al examined the effectiveness of the MTX prophylaxis regimen by examining 585 patients newly diagnosed with DLBCL and at high risk of CNS recurrence, of whom 295 (50%) received MTX prophylaxis, the result suggest that although the risk of CNS recurrence at 1 year was lower than in patients who did not receive MTX (2% vs. 7.1%), this prevention effect was not evident over time. This indicates that MTX is not effective in preventing the onset of CNS, but only has the effect of delaying CNS recurrence or metastasis [20, 21]. Therefore, it is urgent to find an optimal regimen that is effective in preventing CNS recurrence and metastasis. The efficacy of rituximab in preventing CNS recurrence or metastasis remains controversial. Villa, D et al divided patients diagnosed with DLBCL into 2 groups, one with the CHOP regimen and the other with the R-CHOP regimen, and assessed the risk of CNS recurrence in patients on both treatment regimens; the results of the multifactorial analysis showed that the use of rituximab significantly reduced the risk of CNS recurrence with statistical significance (p < 0.05), especially in pathologically in patients in complete remission[22]. Consistently, Schmitz, N et al came to a similar conclusion that the use of rituximab reduced the CNS recurrence rate in DLBCL, especially in younger patients with lower IPI scores[23]. On the other hand, Hosein, P. J et al retrospectively collected data from 76 patients with stage I/II (breast and local lymph node involvement) from eight US academic centers, and after a median follow-up of 4.5 years, the analysis showed no survival benefit with rituximab and the odds of extra-nodal progression in the breast or CNS remained high [6, 24]. CAR-T therapy might be effective in preventing CNS relapse [25].
Hu, S et al concluded from a retrospective analysis of 108 patients with PB-DLBCL that although rituximab reduced the five-year cumulative risk, late recurrence in the contralateral breast and CNS was frequent, and radiotherapy combined with rituximab had a complementary effect on preventing recurrence in patients but was poorly effective in preventing CNS recurrence in PB-DLBCL [26]. Although radiotherapy can control local tumor recurrence, it is not valid for improving patients' OS and PFS; In contrast, a meta-analysis of the published literature on the subject showed that receiving radiotherapy in situ or even contralaterally improved patients' OS and PFS[13, 27, 28].
DLBCL can be classified into GCB type and non-GCB type by immunophenotypic detection [29]. Only 14 patients in our study had this staging, and due to incomplete data, we were unable to perform an accurate analysis of the prognostic differences between the two types of B-DLBCL. According to our knowledge of DLBCL, the non-GCB type has a worse prognosis than the GCB type, and the use of rituximab can eliminate this difference. Wilson, W.H. et al found that adding imatinib to the R-CHOP regimen can further improve the survival of patients with non-GCB type [30–32], but whether this conclusion still holds in patients with B-DLBCL needs further investigation.
In future research, on the one hand, it is necessary to validate and promote the currently considered best treatment options, and on the other hand, it is necessary to search for molecular markers that can help diagnosis and suggest prognosis from the perspective of gene expression profiles, so as to improve patient prognosis in a multidimensional manner in terms of early diagnosis and effective treatment.