Cancer of the Rectum: Neoadjuvant Therapy

Introduction

In the era of total mesorectal excision (TME) surgery, preoperative radiotherapy (RT) reduces local recurrence (LR) and can be administered either as long-course RT (LCRT) with a 5-fluorouracil (5-FU)–based regimen or as short-course RT (SCRT). For patients with a positive circumferential resection margin (CRM) at TME, tumor downstaging is highly desirable. In these circumstances, SCRT is not as effective as LCRT in preventing LR. To reduce radiation-related toxicity, field adjustments and new targeted radiation modalities should be explored.

Radiation therapy was initially introduced as part of the treatment regimen of rectal cancer to improve local control in conjunction with surgery for patients with locally advanced disease. During the past two decades, neoadjuvant (NA) therapy has evolved significantly, not only through its application in serial randomized clinical trials (RCTs) but also as a result of advances in technology that include introduction of computerized imaging techniques. Currently, the treatment of patients with rectal cancer centers around a TME along with NA therapy. Locally advanced rectal cancers are those that are stage III and some that are stage II with extensive local tumor spread. Such patients are treated with similar strategies based on National Comprehensive Cancer Network recommendations ( http://www.nccn.org/clinical.asp ). However, in the era of modern tumor imaging, rectal magnetic resonance imaging (MRI) in particular facilitates the precise estimation of involvement of the CRM, which is the most important factor predicting LR and systemic disease. The Mercury Trial investigators reported that for patients with a positive CRM on preoperative MRI, the risks of LR and systemic spread were significantly higher than for patients with a negative CRM.

Radiotherapy Trials

Surgery remains the cornerstone of rectal cancer treatment for locally advanced (T3/T4) tumors. To further improve local control (LC) in rectal cancer, NA RT has been added to surgical treatment. The benefit of NA SCRT was first shown by the Swedish Rectal Cancer Group in the pre-TME era. After a median follow-up of 5 years, the LR risk was 11% in the group that underwent radiation and 27% in the group that did not undergo radiation, and the overall survival (OS) rates were 58% and 48%, respectively. In spite of these excellent results, the role of RT was questioned after the introduction of TME. In a Dutch TME study, the benefit of SCRT followed by immediate TME was demonstrated, with a 10-year LR rate of 11% for patients treated with TME alone versus 5% for patients treated with SCRT followed by TME. In the MRC-CR07 study, which has a comparable design, patients with resectable rectal cancer were randomized between NA SCRT followed immediately by surgery versus surgery alone. The patients in the group that underwent surgery alone received postoperative chemotherapy (CT)-LCRT when the CRM was involved. The LR rate was 5% in the patients receiving NA SCRT and 12% in patients undergoing surgery alone ( P <.001), demonstrating that selective postoperative CT-LCRT is not able to provide the same results as preoperative treatment.

It has been established that NA RT (either SCRT alone or CT-LCRT) provides significantly better local control than does postoperative CT-LCRT, and LR rates are significantly improved in patients receiving NA SCRT. In this study, the benefits of NA SCRT were observed at all tumor levels, including upper third tumors, and were significant even for patients with specimens obtained within the mesorectal plane. Thus far no survival benefit has been demonstrated for patients who received RT and underwent surgery with the TME technique. In the Dutch TME trial, NA SCRT had no effect on OS or cancer-specific survival when all randomized patients were included in the analyses. However, in patients with a negative CRM who underwent surgery, RT significantly improved cancer-specific survival. Unfortunately, this benefit was offset by an increase in other causes of death, resulting in an equal OS rate compared with the group that underwent surgery only. A subgroup analysis demonstrated that for patients with TNM stage III cancer who had a negative CRM, 10-year survival was 50% in the NA SCRT group versus 40% in the surgery-alone group ( P = .032). In most studies, the LR rate was reduced from greater than 10% for patients treated with surgery only to 5% for patients treated with NA RT and surgery.

NA RT, in conjunction with TME, improves local control but does not improve OS.

Role of Circumferential Resection Margin

In the Dutch TME trial, patients with a positive CRM who had not undergone RT had an LR rate of 23.3%, whereas those who received RT showed a drop in LR to 15.5% ( P = .16). The MRC-CR07 trial showed an LR of 13.8% in patients receiving NA SCRT with 5 × 5 Gy and an LR of 20.7% in patients receiving postoperative CT-LCRT.

In patients with a positive CRM, NA SCRT is not as effective in reducing LR.

Multimodality Treatment Approaches

Several Northern European RCTs have demonstrated a significant reduction in the LR rate after SCRT. Optimal treatment for the different stages of rectal cancer and prioritization of treatment modalities are controversial. In Northern Europe, SCRT is the standard of care for most patients with stage II and III rectal cancer, whereas CT-LCRT is reserved for more advanced cases with a positive CRM. On the other hand, in America and southern Europe, most patients are treated with NA LCRT (LCRT, 45 to 50 Gy) in combination with CT. Furthermore, because the EORTC-22921 and FFCD-9203 studies demonstrated that the addition of CT to preoperative LCRT enhances local control in locally advanced (T3/T4) tumors, with this finding recently confirmed by a recent Cochrane review, patients are now treated with CT-LCRT, followed by TME.

Randomized Trials Comparing Neoadjuvant Short-Course Radiotherapy with Chemotherapy–Long-Course Radiotherapy

Two RCTs compared SCRT to CT-LCRT: a Polish trial, with 312 patients, and an Australian trial, with 326 patients. Both studies had a similar design, and sample size was calculated to demonstrate a difference of 15% in the rate of sphincter preservation and 10% in LR, respectively. Both trials showed higher rates of early radiation toxicity in the CT-LCRT arm when compared with the SCRT group; grade III to IV acute toxicity rates were 18% versus 3% ( P = .001) in the Polish trial and 28% versus 1.9% ( P = .001) in the Australian study. In the Polish trial, the sphincter preservation rate did not differ between the groups, with 61% in the SCRT group and 58% in the CT-LCRT ( P = .57). In this trial, the LR rate was slightly lower in the SCRT group than in the CT-LCRT group at 10.6% versus 15.6% ( P = .21), whereas the opposite tendency was seen in the Australian study at 7.5% versus 4.4% ( P = 0.24). In this latter trial, a difference was observed in the group of tumors at or below 5 cm from the anal verge, with 6 of 48 patients in the SCRT arm versus 1 of 31 patients after CT-LCRT (not statistically significant). In the Polish study, severe late toxicity was observed in 10.1% of patients after SCRT and in 7.1% of patients after CT-LCRT compared with 5.8% of patients after SCRT and 8.2% of patients after CT-LCRT ( P = .53) in the Australian trial ( Table 29-1 ).

TABLE 29-1

Five-Year Local Control and Overall Survival Rate Results from Selected Randomized Clinical Trials Using Preoperative 5-Fluorouracil–Based Chemotherapy and Long-Course/Short-Course Radiotherapy

RCT	T/S Status	No. of Patients	Median Age, yr	LR Rate at 5 yr, %	OS at 5 yr, %	pCR Rate, %
German trial: preoperative	T1-T4	405	62	6	76	1.9
German trial: postoperative	T1-T4	394	62	13	74	0 ( P < .001)
Dutch trial TME alone	S1-S4	908	66	10.9	63.5	0
Dutch trial NA SCRT + TME	S1-S4	897	65	5.6	64.2	0
MRC-CR07 NA SCRT	S1-S4	674	65	4.7	70.3	N/A
MRC-CR07 postoperative CT-LCRT	S1-S4	676	65	11.5	67.9	N/A
Polish trial SCRT	T3-T4	155	60	9	67.2 (4 yr)	0.7
Polish trial CT-LCRT	T3-T4	157	59	14.2	66.2 (4 yr)	16.1 ( P N/A)
Trans Tasman SCRT	T3	162	63	7.5 (3 yr)	74	0
Trans Tasman CT-LCRT	T3	161	64	4.4 (3 yr)	70	12.4 ( P N/A)
FFCD-9203 NA LCRT	T3-4	367	63	16.5	67.9	3.6
FFCD-9203 NA CT-LCRT	T3-4	375	64	8.1	67.4	11.4 ( P < .0001)