The relationship between post-surgery infection and breast cancer recurrence

Predominantly a disease that presents in females, breast cancer has been identified as the most commonly diagnosed form of cancer in women globally, with an estimated 2 million reported cases a year []. Breast cancer for the most part is a treatable disease, with the survival rate improving due to enhancements in screening and treatment []. However, cancer recurrence remains a dominant contributor to breast-cancer-related deaths. Specifically, this is when a subpopulation of primary tumour cells acquire genetic and epigenetic changes and may either persist as dormant or spread systemically, evading treatment, and facilitating relapse months or years later. When breast cancer recurs, the 5-year survival rate can drop from 70–80% to less than 30% []. Breast cancer recurrence can occur as either local recurrence (LR), at the same site, or as distant recurrence (DR), metastasis to a different anatomical site, or both. The risk of 10-year recurrence varies depending on cancer subtype and adjuvant therapy, but has been found to range from 4% to 34% [].

Surgery remains the standard treatment for breast cancer. However, the adverse impact of surgery remains controversial, although not yet fully understood. It has been suggested that the change in tissue and tumour microenvironment (TME) caused by the surgery may alter the growth kinetics of breast cancer cells []. Surgical site infections (SSIs) have been shown to be the leading cause of operation-related adverse events [,]. They represent a significant issue that can reduce the quality of life and prognosis of patients following surgery, as well as increase the financial burden to patients, hospitals and governments. In the USA, Olsen [] attributed an incremental cost of over $4000 per patient in the event of an SSI.

Breast surgery is classified as a clean surgical procedure, with no exposure of the respiratory, alimentary or genitourinary tracts. As such, the expected rate of SSI incidence in the postoperative period is approximately 1–3% []. However, studies examining this have found that this approximation does not hold up in practice, suggesting that such procedures should be treated as clean-contaminated. Previous systematic literature reviews found that the use of antibiotic prophylaxis, not normally required for clean surgeries, reduces the likelihood of obtaining an SSI []. Indeed, breast cancer may be a key variable associated with SSI. Infection rates are higher in patients with breast cancer when compared with non-cancer patients who undergo similarly extensive operations, such as breast enhancement surgery []. Recent studies have also demonstrated that the breast tumour microbiome is distinctly different than that of the normal breast tissue, with a more rich and diverse bacterial load. Interestingly, they also demonstrated that breast tumours have larger and more diverse microbiomes than any other tumours they examined [].

The progression, treatment and prognosis of breast cancer is influenced considerably by the TME. The TME in breast cancer is a complex structure comprising of stromal cells, including fibroblasts, mesenchymal stromal cells, osteoblasts, adipocytes and pericytes as well as non-stromal cells, such as the extracellular matrix (ECM) and immune cells [,]. Disturbances of the TME, such as SSI at the time of primary surgery, have been correlated with breast cancer recurrence [,], but the mechanisms mediating this phenomenon have yet to be elucidated. Theoretically, infections may promote a local immune-derived anti–tumour response. Bacterial infection may potentially stimulate the host's natural killer cells and macrophages, inducing a strong anti-tumour immune response protecting against invading pathogens and transformed cells, including cancer cells []. Conversely, it has been suggested that, as acute infections stimulate local levels of cytokines, growth factors, and proteinases, these may in turn alter the TME to support tumorigenesis. Breast cancer cells, influenced by their microenvironment, ‘hack’ the tumour promoting cancer-associated fibroblasts and cytokines, increasing proliferation, migration and invasion [,].

While there is emerging evidence that occurrence of SSI may relate to breast cancer recurrence, the influence that these bacterial infections have on breast cancer recurrence rates and the mechanisms required to enable cancer cells to reoccur in the presence of infection, have yet to be determined. Consequently, a more comprehensive understanding of SSIs and their causative pathogenic microbes is critical to understanding their potential influence on the TME.

Therefore, the aim of this review was to critically appraise current research regarding the role that SSIs and post-primary breast cancer surgery play in wound healing and breast cancer recurrence, and the plausible mechanisms associated with such interaction.

This was performed using two approaches: the first objective was to explore SSIs after primary breast cancer surgery and the second aim was to examine cases of breast cancer recurrence following primary breast cancer surgery, where occurrence of SSIs were also recorded, and to delve into potential links between these two events.

This review addressed two principal areas: first, establishing the characteristics recorded regarding SSI after primary breast cancer surgery; and second, examining the events of breast cancer recurrence after SSI of the primary breast cancer surgical wound and the plausibility that they are linked.

The first aspect of this review extracted data from 484,605 patients in the 99 eligible studies (Figure 1) and the second, 17,569 patients from 53 studies. The results of this review suggest that the association between postoperative infection and adverse cancer outcomes is clear, but the cellular link between them remains elusive. There is evidence that specific bacterial species are associated with SSI from the breast surgical wound of the cancer patients. S. aureus, E. coli and P. aeruginosa were the most common causative bacterial species identified, consistent with emerging literature []. There was a high proportion of DRs. Five studies investigated the association between SSI and breast cancer recurrence with three concluding that an association did exist. Arising from this review, albeit that confounding factors make results challenging to compare and interpret, there are sufficient data to reasonably argue that a standardized prospective study with appropriate statistical power is warranted.

The SSI incidence rates reported in individual studies are quite variable, ranging from 0.2% to 48%. The majority of these were higher than the 1–3% suggested SSI rate for clean surgeries of the breast []. There are no obvious explanations for these inconsistencies in infection rates between institutions. Previously, it had been suggested that staff training, the experience of surgical and perioperative teams, and use of antimicrobials prophylactically may all contribute to these discrepancies [,]. Perhaps it proposes that the cancer itself may be playing a role in the development of these infections. This theory is supported by Olsen [] who examined both cancer and non-cancer patients undergoing breast surgery and found that the cancer patients had a higher incidence of SSI despite the fact that they have very similar surgical procedures in terms of duration and invasiveness. Interestingly, a recent study by Nejman [] found that the bacterial load of the tumour microbiome was much larger than that of the normal breast samples, this was not the case for other cancers, such as lung and ovarian.

There was considerable variation in criteria used to define SSIs throughout the studies included in this review. Currently, there is no worldwide ‘gold standard’ classification method, making interpretation of the incidence rates challenging. Table IV demonstrates how four different definitions of SSI were used, but nearly 50% of studies did not mention whether they employed a specific definition. Notably, a study performed in the USA, found that by changing their definition of what constituted an SSI after breast surgery, by excluding cases of cellulitis, the incidence rate dropped from 8.7% to 2.7% []. Moreover, the retrospective nature of the majority of the studies may have influenced the SSI rate recorded, possibly resulting in higher rates due to follow-up being conducted on an outpatient basis. If an SSI developed but did not require readmission then it may not have been documented in the patient records, meaning it was not included in the study [,]. Almost one-third of the studies had a follow-up period of one month, owing to the fact that an SSI is typically defined as occurring within the 30-day time period after surgery []. However, in the latest guidelines published by the Centres for Disease Control and Prevention (CDC) they advise that a 90 day surveillance period for surgeries of the breast should be employed []. It is also worth noting that many of the papers did, in fact, monitor the patients postoperatively for much longer than the 30 day period suggested by the definition and, consequently, recorded a higher incidence of SSI than those only followed for one month. The mean rate of infection increased from 11%, for those monitored for 30 days, to 13.4% for those with a longer than 30 days follow-up period. However, it should be noted that more standardization across studies is needed in order to determine if this is a true difference. Other studies have found that it may be necessary to extend the surveillance to 265 days post-surgery in order to fully detect all SSIs [, , ]. Overall, this inconsistency in the numerous SSI definitions employed and capricious data-collection methods acts as a confounder of results, making it challenging to compare variables across studies.

The microbiome of breast tissue has normally been associated with Gram-positive phyla such as staphylococci species [,] but more recently it has been suggested that the epidemiology of the breast, especially breast tumours, is changing with more Gram-negative bacteria being identified in surgical wounds []. The results of this review support this, as it identified a much larger than expected volume of infections caused by Gram-negative bacilli, namely, E. coli and P. aeruginosa (Figure 2). This may be suggestive of poor hygiene and cleaning practices contaminating the surgical field or the wound itself. The method of obtaining cultures was not standardized; as such, some may have been obtained under sterile conditions, and others may have been superficial swabs. Methodological confounders may be reflected in studies whereby, for instance, all of the 43 reported cases of E. cloacae species were found by the Vilar-Compte group in Mexico [, , ]. Similarly, Rolsten et al. from the USA are unique in reporting Proteus mirabilis from the SSIs of the breast cancer patients in their institution in Texas [,]. This is suggestive of environmental factor or testing-related confounders that contribute to the outcomes of their studies.