ABSTRACT: The optimal approach to the diagnosis and treatment of locally advanced rectal cancer involves multidisciplinary, integrated management. In the past 30 years, survival and freedom from disease have increased, but the ideal multidisciplinary management remains to be determined. The preferred integrated treatment modality is preoperative radio(chemo)therapy followed by total mesorectal excision. Certain aspects of this standard are still debated, and the European and American approaches vary. The chief recommendations per international guidelines are summarized, and the next generation of integrated treatments for locally advanced rectal cancer is discussed.
Near circumferential mid-rectal cancer seen on colonoscopyWorldwide, rectal cancer is a leading cause of death from cancer, affecting males and females equally.[1-3] In the United States, there were 39,870 new rectal cancer cases in 2011.[4]
Since the 1980s, the outcomes of rectal cancer have improved; both survival and disease-free intervals have increased through advances in diagnostic imaging, pathology, surgery, and integrated treatments, including radiotherapy (RT) and chemotherapy (CT). More important, the integration of these different treatment modalities has led to the modern standards for rectal cancer treatment. Here, we summarize the principal aspects of current standards for integrated treatment of locally advanced rectal cancer (LARC), comparing the European and American approaches.
Must preoperative radio(chemo)therapy be added to optimal surgery?
Surgery is the historical cornerstone of rectal cancer treatment. In the 1970s, unsatisfactory results were associated with surgery alone; local failure rates were 50% for locally advanced and node-
involving presentations,[5,6] prompting attempts to increase local control and survival.
In the 1980s, adjuvant postoperative concurrent radiochemotherapy (RTCT) was evaluated in North America and was found to significantly improve survival and time to recurrence.[7,8] In 1990, the National Institutes of Health (NIH) Consensus Conference recommended postoperative integrated RTCT as the standard of care.[9] Around the same time that the NIH consensus statement was issued, preoperative integrated treatments were developed in Europe.[10] The 1997 Swedish Rectal Cancer Trial[11] reported significant improvement in 5-year overall local control and survival with preoperative short-course (SC) RT (5 Gy per 5 fractions, total 25 Gy) followed by surgery, vs surgery alone. Most initial findings on the benefits of adjuvant RTCT in both the preoperative and postoperative settings came from series in which patients received what we now consider inadequate surgery.
The evolution of surgical techniques in European studies culminated in total mesorectal excision (TME), first promoted by Heald, wherein en bloc resection of the rectum with its mesorectum is performed to the levator muscles in the avascular plane outside of the mesorectum. The mesorectum is enveloped by the visceral rectal fascia, which separates it from the parietal endopelvic fascia; sandwiched between the two fasciae lies a thin layer of fat that is filled with loose tissue (the “holy plane of Heald”), containing nerves for the pelvic organs. Embryologically, the rectum and mesorectum (and their vessels and lymphatics) develop together; thus, the plane that describes this resection respects the anatomy and embryological origins.[12,13]
This achievement became the new standard: local failures after TME alone for pT3-4 and N1-2 disease with a medium-risk or low-risk presentation ranged from 15% to 21%,[14] and 5-year local recurrence reached as low as 5%, causing integrated approaches with an adequate and meticulous surgical procedure to be questioned.[15]
European studies concluded that surgical quality significantly influences local recurrence and survival. Quirke proposed three planes, classifying the quality of the respective resected specimens as high, acceptable, or poor: mesorectal plane (intact mesorectum—high quality), intramesorectal plane (minor irregularities of the mesorectal surface—acceptable), and muscularis propria plane (specimen with high loss of mesorectum—poor quality). This classification correlates significantly with survival outcomes.[16]
Studies in the mid-1980s demonstrated that margin involvement was central in effecting local control and that the distance between microscopic tumor penetration and surgical circumferential resection margin (CRM) was more significant than the distal or proximal margin.[17] European groups have recently reported that magnetic resonance imaging (MRI) predicts the risk of CRM involvement or threat before surgery. MRI can define involvement of the mesorectal fascia (MRF); if the MRF is involved or if the lesion gets as close as 1 mm from the MRF, there is a high risk of a positive CRM if only TME is performed—hence, the proposal of the term “MRF +/−”[18,19]
The “Dutch Trial” (2001) was a multicenter randomized study with certified quality of surgery; it compared TME alone to preoperative SC RT followed by TME. Of the 1861 subjects, local failure rates were better in the integrated treatment arm for stage II and stage III patients than in the surgery-alone arm: 2-year overall local failure was 2.4% and 8.2%, respectively (P < .001) (stage II: 1% vs 5.7%; stage III: 4.3% vs 15%). Survival did not differ (82.0% vs 81.8%).[20] This study supported integrated treatment, even if adequate and certified TME was performed. In the 12-year follow-up update, the 10-year cumulative local recurrence rate was 5% for RT plus surgery and 11% for surgery alone(P < .0001). Overall survival did not differ between the two groups. Notably, in TNM stage III patients who had a negative CRM, 10-year survival was 50% in the RT-plus-surgery group compared with 40% in the surgery-alone group (P = .032).[21] The “Dutch Trial” concluded that a positive CRM was a robust predictor of local recurrence, for which planned postoperative RT, in patients with pathological evidence of CRM positivity, could not compensate.[22]
In 2009, a British group examined whether selective postoperative chemoradiation improved outcomes in the subset of patients with involved or threatened CRMs. The UK Medical Research Council Trial (MRC 07) randomized 1350 cStage I-III patients to preoperative SC RT followed by surgery or to initial surgery followed by postoperative long-course RTCT (with RTCT only for patients with involved or threatened CRMs after surgery). TME was not mandatory but was performed in 92% of cases. At a median of 4 years of follow-up, the preoperative arm had significantly lower local failure rates and better disease-free survival (DFS).[23] In patients with a positive CRM, local recurrence rates did not differ significantly between groups. This trial did not have as an aim comparing SC and LC treatment; rather, it compared the preoperative and selective postoperative approaches.
A German group examined the efficacy of preoperative RTCT (the “German Trial”), randomizing 823 patients with cT3 or cT4 or node-positive disease. One arm received preoperative long-course RTCT (LCRTCT) followed by surgery (with mandatory TME), and the other arm underwent the same surgery, followed by postoperative RTCT; the same RTCT fluorouracil(Drug information on fluorouracil) (5-FU)-based regimen (1.8 Gy per fraction, total 50.4 Gy) was administered, except for a boost of 540 cGy in the postoperative arm. The 5-year local failure rate was lower in the preoperative therapy arm (6% vs 13%; P = .006), as were acute and late toxicity rates. Five-year survival did not differ significantly.[24] The 11-year follow-up update confirmed these results.[25]
Based on this and subsequent evidence, the 1990 NIH guidelines are no longer followed in the US, and the postoperative approach is limited to patients with pathological evidence of high risk or who are unable to receive RT preoperatively. Thus, the preoperative RT approach is now preferred, even if adequate surgery is performed; this can be administered in an “SC RT alone” schedule or in LCRTCT (taking advantage of chemosensibilization). It improves local control and (primarily with LC administration) effects pathological complete responses (pCRs) and sphincter preservation, although it has not affected survival in any randomized study in the TME era.
The use of MRI preoperative parameters to tailor treatment can exploit the potential of LCRTCT to manage a positive MRF or a low-lying lesion, as LCRTCT can enhance downstaging more than SC RT.
Five meta-analyses have reported conflicting results on survival.[26-30] All demonstrated a decrease in local recurrence rates with preoperative irradiation. Camma et al[26] and the Collaborative Colorectal Cancer Group[27] reported a survival advantage for the use of preoperative RT. A systematic review of radiation therapy trials by the Swedish Council of Technology Assessment in Health Care (SBU)[28] noted that preoperative RT improved survival by approximately 10%, whereas meta-analyses by Munro and Bentley[29] and Fiorica and Cartei[30] did not. A pooled analysis of post-2000 randomized trial data showed a survival benefit for preoperative RT.[31]
