Mouse model of plasma cell mastitis

Plasma cell mastitis (PCM), also known as granulomatous mastitis or nonpuerperal mastitis, was first proposed by Cheatle and Cutler, in 1931[1]. It was a nonbacterial inflammatory breast disease frequently occurred among young and middle-aged women at non-pregnancy or non-lactation stages [2, 3]. According to the current report, the incidence rate of PCM increased gradually, and extended to pubertal or menopausal women [4‐7]. However, little is known about its mechanism as its onset is occult, but several studies have shown that recurrence of PCM was commonly observed [8‐10]. Surgical excision is the major treatment of the disease [2, 10, 11] and wide local excision was commonly performed as limited surgical excision might lead to higher rate of recurrence [10]. However, the operation could not prevent the recurrence of the disease as a discrete mass [12, 13] even though the mammary gland was removed, thus women might experience mental and physical distress.

The inflammatory disease clinically mimics breast cancer [14, 15] and some cases were misguided as breast cancer [16, 17]. Histologically, the disease is presented with duct widening characterized by periductal inflammation and fibrosis. Some characteristic calcification can be seen by mammography. Microscopically, granuloma formation, Langhan’s giant cell, plasma cell, histiocyte, lymphocyte, and caseous necrosis were observed [18, 19]. However, a prospective follow-up study of 270 patients with nonpuerperal breast inflammation has demonstrated an increased risk of breast cancer in women with nonpuerperal mastitis [20]. The correlation between the inflammatory disease and breast cancer has not yet been established.

In this study, a PCM mouse model was established by inoculation of aseptic homogenate of human PCM pathological tissues and complete Freund’s adjuvant (CFA) to investigate the immunologic mechanism of PCM.

PCM is a rare breast disease that mimics breast cancer and the etiology remains unclear. It is therefore clinically significant to differentiate the inflammatory disease from malignant disease. Basically, the disease could be categorized into infectious and non-infectious ones which respectively presented as tuberculous mastitis and idiopathic granulomatous mastitis with distinguishable histological features [4]. However, the immunological mechanism of PCM and its associated pathological change were still unknown. A variety of different terms including granulomatous mastitis, mammary duct ectasia or nonlactational mastitis created unconformity of the designation systems. Previous studies showed anaerobic infection was the major cause of PCM, which might result in tumorigenesis in the infected area [21, 22], but the casual relationship between mastitis and breast cancer was not fully supported and widely accepted [23]. Currently, PCM was reported to be associated with nonbacterial infection [24]. Our data indicated that most of the PCM patients showed a history of inverted nipples, which might lead to stenosis and obstruction of mammary duct, as well as deposition of secretions such as breast milk in the mammary duct. We speculated that long-term deposition of breast milk in mammary duct could stimulate autoimmune response, followed by pathological changes in different parts of the mammary gland [3].

Surgery was still the preferred treatment of PCM to the antibiotic treatment which was not efficient. However, complete remission was not guaranteed after surgery. To date, bacterial mastitis mouse models were mainly induced by injection of bacterial liquid via lacteal gland [25] and in a retrospective review of 26 clinical cases of granulomatous mastitis, idiopathic cases were not associated with specific micro-organism with negative special stains and cultures for micro-organisms [26]. In this study, a PCM mouse model was induced by active ingredient of the PCM tissues. The possibility of bacteria induced PCM was eliminated by sterile processing from sample collection to inoculation and ultrasonic cell disruptor processing after homogenization. Moreover, CFA was used to promote the formation of PCM due to the fact that CFA could function as antigen reservoir and stimulate local inflammation and granulomatous reaction as well as the proliferation and differentiation of lymphocytes [27]. The use of CFA in this study could preclude the possibility of positive pathological changes of PCM resulted from enhancement of immunological responses to antigens.

PCM mouse models were established in group C and group D. Infiltration of plasmocytes, neutrophils and lymphocytes was detected in peripheral breast tissues, which was consistent with the diagnostic standard of PCM. No statistical difference of PCM pathological changes was observed between groups. Also, manifestation of PCM was not observed in group A and B, demonstrating that physiological saline did not have any effect on the development of PCM. Infiltration of neutrophils and lymphocytes was detected in group E, indicating that CFA was not efficient in the induction of PCM even though it functioned in promoting inflammation. We therefore confirmed that certain ingredients of pathological tissues from PCM patients could induce the development of PCM in mouse models. The first PCM mouse model presented in this study will provide a platform for the investigation of the etiology, pathological mechanism and the potential drug selection for the PCM.

The PCM lesion from patients could induce the development of PCM in female BALB/c mice with similar clinical and pathological manifestation and higher dose of PCM homogenate succesfully developed PCM in all mice. The mouse models could be futher investigated to explore the etiology, disease mechanism as well as potential drug therapy while at the moment traumatic surgery is still the major treatment for patients with PCM.