The unpleasant symptoms from CWR, such as lymphoedema, ulceration, pain, fungal infection and persistent disease progression pose challenges in the clinical practice. However, these patients represent a heterogeneous population, and some of them may achieve a long-term survival after CWR. Here we hope to summarize the characteristics of breast cancer patients with CWR and provide novel insights for patient stratification and further individualized treatment. In the present study, we found that skin involvement predicted poor local disease control in breast cancer patients with CWR, especially local disease progression events. In addition, patients with skin involvement tended to experience persistent chest wall progression after CWR than those without skin involvement.
The mechanisms that predispose certain internal malignancies to invade the skin have rarely been discussed in the previous literature. Skin can be divided into epidermis and dermis in anatomic structure. A complex network of multitude of cell types and the flexibility of dermal vessels and the lymph nodes in skin maintain several vital processes such as inflammation, immune response, wound healing, and angiogenesis [20–24]. The levels of tumor-infiltrating lymphocytes (TILs) in skin metastases are the lowest and the composition of TILs presents a higher FOXP3 and a lower CD8/FOXP3 ratio, compared to other metastatic lesions from primary breast cancer [25]. This suggests that cutaneous tissue may harbor a more permissive immune microenvironment for tumor growth. A physiological mitigation of the cytotoxic immune activity in skin tissue through different immunosuppressive mechanisms, a process known as “immune privilege”, has been described [26, 27]. In addition, the interaction between tumor cells and certain factors secreted from the dermis or epidermis participate in the skin homing mechanism of metastatic cells. Chemokines and their receptors are involved in tumorigenesis and metastasis. The chemokine receptor CCR10 has been demonstrated to be involved in cutaneous metastases of melanomas and may mediate melanoma survival in the skin [28, 29]. Breast cancer, which has a similar metastatic pattern as malignant melanoma but also a high incidence of skin metastases, shows high expression levels of CXCR4 and CCR7 rather than CCR10 [30]. Structurally, skin involvement from breast cancer of female occurs mainly by hematogenous and lymphatic routes [18, 31]. Regional distribution of dermal lymph vessels, from the perspective of tumor invasion, constitutes a functional unit [32]. Joan et al. finds that more superficially located tumors may be more likely than are deeper tumors to metastasize via lymphatics to axillary nodes due to the rich lymphatic system in the breast dermis, although the size of primary site is less than 5 cm [33]. In addition, cutaneous deposits from breast adenocarcinomas with distinct angiogenic profiles in skin have different growth patterns [34]. The suppressive immune microenvironment in the skin lacks supervision and killing of tumors, and the abundant vascular and lymphatic system in the dermis provides nutrition and spreading channels, which contribute to the growth, invasion and metastasis of tumors. This may partly explain why skin involvement is an independent prognostic factor for disease progression in breast cancer patients with CWR, especially local disease progression.
The treatment of recurrent chest wall disease is complex, particularly when radiotherapy has been previously conducted for chest wall. Full thickness chest wall resection, as one of the most common therapeutic approaches, is not curative, is disfiguring, and is associated with significant morbidity. Previous studies suggested that multiple sites or large areas of skin involvement are not suitable for local surgical treatment, although systemic chemotherapy is also difficult. Second-and third-line salvage chemotherapy results in overall response rates of 20–30%, at best [35–38]. In our study, skin involvement predicted poorer local disease control in patients with CWR, while similar overall survival was observed between these patients with skin involvement or not. Our results suggested that skin involvement should not be considered as a contraindication for local therapy in breast cancer patients with chest wall recurrent diseases. On the contrary, more effective treatment strategies and more clinical studies are needed to improve clinical outcomes and quality of life for these patients, especially in the relatively early stage of skin invasion.
A certain proportion of patients with CWR after local or systemic treatment have second recurrence or disease progression in a shorter time. However, biomarkers for subsequent recurrence or disease progression were not consistent. Anees et al. finds that primary tumor characteristics and treatment factors, time to recurrence, characteristics and treatment of recurrent chest wall were not significantly predictive for an additional CWR in their study [16]. The study from Carmen et al. shows that disease free interval between primary lesion and CWR remain highly correlated with disease progression after chest wall chest [14]. Bruce G et al. demonstrates that PR status is a significant factor for distant metastasis and HER-2/neu status is related to local-regional disease progression after local recurrence [6]. The varying lengths of follow-up, different patient selection criteria, and a variety of treatment strategies can be an explanation for the difference among varied studies. Although some studies explore the second CWR after CWR, that is, re-recurrence, there are few studies on persistent chest wall progression. To our knowledge, this is the first study on the characteristics of patients with persistent chest wall progression and the prognosis of skin involvement in these patients. In our study, initial nodal status, status of ER, PR and HER2 and skin involvement are significantly different between patients with persistent chest wall progression and non-persistent chest wall progression. Persistent chest wall progression was likely more to be associated with positive lymph node status, negative PR and positive HER2 in primary lesion, negative ER and PR in chest wall lesion and skin involvement of chest wall. Positive lymph node status, negative HR status and positive HER2 status were generally markers of poor prognosis in breast cancer. KI67, as a nuclear protein expressed in proliferating cells, is associated with poor prognosis of breast cancer. Consistent with our results, KI67 expression was also elevated in recurrent chest wall lesions and was associated with higher aggressiveness.
The current study had several limitations. Studies involving multiple institutions would add more power to the study. Like any retrospective analysis of this nature, the population was accrued over 2 decades, during which treatment for breast cancer as well as treatment for recurrences changed based on new guideline. Not all patients received the same treatment, even if there was no significant bias between basic therapy, such as systemic therapy and local treatment. Acknowledging the limitations of such an analysis, however, we believe the results can help individualize treatment decisions and provide a strong incentive for future investigations. Certainly, patients with CWR without skin involvement will enjoy a relatively high rate of local disease control and a long-term progression free survival.