Advancing the surgical treatment of locally recurrent rectal cancer

Local recurrence of rectal cancer occurs in 5–15 per cent of patients1. Without treatment, survival is poor, with a median survival of 6–7 months frequently associated with symptoms of chronic pain, pelvic sepsis, fistulas, obstruction and bleeding2. Many patients suffer a miserable death from their malodorous, fungating tumour. Radiotherapy, alone or in combination with chemotherapy, can palliate symptoms and extend survival to 12–14 months3. Radical surgery is effectively the only curative option in this patient cohort. Radical surgery for locally recurrent rectal cancer (LRRC) was first conceptualized in the 1940s2. Since then, a number of series have described oncological and survival benefits with 5-year survival rates of 38–62 per cent4–6. Despite such promising outcomes, only a small proportion of patients are referred for consideration of curative surgery. Unlike the mainstream acceptance of surgery for hepatic and pulmonary metastases, general recognition that surgery for local recurrence can offer benefit has been slow. Five-year survival rates following hepatic and pulmonary metastasectomy are 34–44 and 39–48 per cent respectively. Survival benefit after surgery for local recurrence is comparable, if not superior, to that for metastasectomy, and therefore such intervention should be considered seriously for patients with local recurrence. A recent UK-based survey of colorectal surgeons suggested that about 700 patients had been managed with LRRC over a 12-month period with only four units operating on more than ten patients a year and no concordance in management, reflecting the lack of guidelines on this subject7. Reservations regarding surgery for LRRC may be due to the historical perception of the associated high morbidity and mortality, with reported rates of up to 59 and 3 per cent respectively8. Much of this morbidity is related to the perineal wound, with high rates of wound dehiscence, infection and delayed healing attributed to the poor quality of previously irradiated tissue. Similar high perineal morbidity rates have been observed with abdominoperineal resection for primary lower rectal cancer. Recent advances in perineal reconstruction using myocutaneous flaps and biological mesh have provided encouraging results with a reduction in perineal morbidity9. Advances in surgical practice have expanded the range of options available to patients with LRRC, with the choice of surgical procedure being dictated by the anatomical location of the tumour. Typically, there are three common sites of recurrence: central or anastomotic recurrences, lateral recurrences involving pelvic side-wall structures, and posterior recurrences involving the presacral space or the sacrum itself. Accurate delineation of the anatomical position and involvement of surrounding structures using magnetic resonance imaging (MRI) with gadolinium enhancement and computed tomography–positron emission tomography (CT–PET) is important to provide a road map for surgery. MRI accurately predicts adjacent organ involvement, and can differentiate between fibrosis and disease recurrence by the rim enhancement that is displayed by recurrent tumour following gadolinium administration10,11. CT–PET combines anatomical and functional data by the uptake of [18F]fluorodeoxyglucose to define the presence, localization and extent of the recurrence, and the detection of occult metastatic disease12. The anatomical position of local recurrence not only dictates operative management but also has significant implications for the radicality of resection and therefore on prognosis, with central recurrences most likely to achieve R0 resection status and consequently superior survival benefit compared with posterior and lateral recurrences4,5. There are a number of contraindications to curative surgery. High involvement of the sacrum above the level of S3, extensive involvement of the lateral pelvic side-wall, extension through the greater sciatic notch and encasement of the iliac vessels13 are all considered unlikely to result in R0 resection without considerable postoperative morbidity, significant disability and impaired quality of life. High sacrectomy14, external hemipelvectomy15 and en bloc resection of the iliac vessels16 have, however, all been performed with curative intent and fewer problems than anticipated. Some 89 per cent of patients undergoing a high sacrectomy were able to ambulate independently after surgery or with the assistance of a walking aid, and all patients reported acceptable postoperative quality of life as defined by resumption of activities of daily living and reduction of cancer-related pain14. Half of the patients following an external hemipelvectomy were able to walk with the aid of a prosthesis15. Surgical advances have been augmented by other therapies, including intraoperative radiotherapy and selective re-irradiation. Further irradiation using external-beam radiotherapy with doses of 30–40 Gy has been given with acceptable rates of toxicity in order to increase the likelihood of achieving an R0 resection margin in combination with radical surgery. Intraoperative radiotherapy in the presence of a positive resection margin can also be used to improve local control. Surgery for LRRC is technically demanding because of the loss of normal anatomical planes, postradiotherapy fibrosis and invasion. Operations are long, blood loss can be high and the involvement of a number of allied disciplines calls for the development of specialist, centralized services to assess and manage these patients. This should lead to better selection of patients, more accurate preoperative staging of disease, optimization and standardization of multimodal therapies as well as facilitating the refinement of surgical techniques. It is well documented that surgical outcomes are improved by high-volume surgeons in high-volume centres. The generation of evidence-based clinical guidelines, meaningful research and specialist training are all consequences of such rationalization leading to better care for these patients in the future. Just as metastatic disease is no longer necessarily the end of the road, neither is local pelvic recurrence. LRRC is a potentially curable disease process. Perhaps the mindset towards the management of LRRC needs to change, with a shift towards the consideration of radical surgical resection in conjunction with other therapies. The surgical community needs to adopt the attitude that, as with metastatic disease, pelvic recurrence is not necessarily terminal. The authors declare no conflict of interest.