In recent years, multimodal treatment including initial preoperative chemoradiotherapy (CRT) followed by total mesorectal excision has become the preferred treatment for advanced rectal cancer, and this strategy is widely accepted as a standard treatment in western countries (1). Randomized clinical trials have showed that fluorouracil-based preoperative CRT reduces local recurrence compared with preoperative radiation without chemotherapy or postoperative CRT (2-4). Conventional intravenous 5-fluorouracil (5-FU) or capecitabine have been used as radiosensitizing agents in CRT in recent years (1). However, further improvements in this therapy are needed in order to achieve better outcomes of cancer treatment and a better quality of life for patients. Therefore, different chemotherapeutic regents and regimens as radiosensitizers have been recently attempted for CRT (5).

S-1, a biochemical modulator of 5-FU, is an oral dihydropyrimidine dehydrogenase inhibitory fluoropyrimidine containing tegafur, gimeracil (5-choro-2,4-dihydroxypyridine) and potassium oxonate in a molar ratio of 1:0.4:1(6). Tegafur is a prodrug of 5-FU, and gimeracil is a competitive inhibitor of dihydropyrimidine dehydrogenase that degrades 5-FU (6). S-1 contains not only tegafur as a radiosensitizer but also gimeracil, which enhances the efficacy of the radiotherapy (7-9). However, few studies regarding preoperative CRT with S-1 for rectal cancer have been conducted thus far (10-12).

The present study therefore investigated the short-term outcomes of neoadjuvant CRT with S-1 for locally advanced rectal cancer.

Preoperative CRT is a key treatment for enhancing the local control and reducing the therapeutic toxicity compared with postoperative CRT. Preoperative CRT with continuous infusion of 5-FU or capecitabine is the standard treatment for rectal cancer (1,13). 5-FU-based reagents function as radiosensitizers by killing S-phase cells, which are relatively radioresistant (14). Recently, capecitabine has been used to replace treatment with 5-FU as neoadjuvant chemoradiotherapy for locally advanced rectal cancer (13). However, a previous randomized study showed that CRT with capecitabine was associated with a significantly high incidence of hand-foot skin reaction (31%; any grade), fatigue (28%; any grade) and proctitis (16%; any grade) compared with 5-FU (13). In contrast, S-1 is also used for adjuvant therapy after rectal surgery and is considered quite tolerable (15). The oral anticancer reagent S-1 is a biochemical modulator of 5-FU and has been reported to enhance the effect of radiotherapy in cancer cells (7). Therefore, preoperative CRT with S-1 may be a viable candidate for the treatment of the rectal cancer.

In the present study, the frequent toxicity of this treatment was anal pain (69%), followed by skin and subcutaneous tissue disorders (59%). These toxicities are notable findings because no similar findings had ever been mentioned in previous reports (10,11). However, it was a reversible event, and all patients recovered after the treatment before radical surgery. When performing chemoradiotherapy with S-1, we need to keep in mind the possible occurrence of skin related side effects. Regarding severe toxicity, grade 3/4 leukopenia, anemia, a neutrophil count reduction, a platelet count reduction and diarrhea were seen in 6, 3, 3, 3 and 3% of cases, respectively. These results were comparable to those of previous studies of preoperative CRT with 5-FU or capecitabine (leukopenia: 8.2 and 1.5%, anemia: 1 and 0%, platelet count reduction: 0.5 and 0, diarrhea: 2 and 8.6%, respectively) (13). One patient terminated treatment due to severe diarrhea. However, completion of this combined therapy without any modification was seen in 93.8% of cases of S-1 treatment and 96.9% of those with radiotherapy. Indeed, preoperative CRT with S-1 is reported to have a high compliance rate for chemotherapy (89.2-100%) and radiotherapy (94.6-100%) (10,11), with two reasons speculated: Oral treatment can be administrated on an outpatient basis, and S-1 is less toxic than intravenously delivered chemotherapeutic reagents (11).

A previous study showed that the pCR rate of CRT with S-1 was 10.8-22.2% (10,11). Our study showed a pCR rate of 28.1%, which was consistent with the findings of previous reports. Capecitabine, an orally administrated pro-drug of 5-FU, is already widely used in western countries, and the pCR rate of CRT with capecitabine reportedly ranges from 12-31% (16-18). In addition, the pCR rate of the floropyrimidine (5-FU, capecitabine, S-1)/irinotecan, floropyrimidine (5-FU, capecitabine)/oxaliplatin and floropyrimidine (5-FU, capecitabine)/bevacizmab with radiotherapy regimens for have been reported to be 13.7-37, 13.3-19.2 and 7.5-32% and respectively (5). These results are comparable to those of CRT with S-1, and the tolerability of this therapy seems acceptable, suggesting that this regimen should be considered in the future.

In our study, R0 resection was performed in 100% of patients. A previous report on CRT with S-1 also found a similarly high R0 rate (94.6%) (11). However, anal-preserving surgery was only performed in 9 patients (28%) in our study, which seemed relatively low compared with the previous report of Hiratsuka et al (12) (43.2%). Twenty-eight of 32 tumors were located in the lower rectum. Therefore, the anus could not be preserved in 19 patients with lower rectal cancer because a safe surgical margin could not be achieved due to the fact that the tumor was located close to the anus, and/or defecation dysfunction was deemed likely to occur due to the patient's age and background. On the other hand, among the 9 patients who achieved a pCR in whom the creation of a colostomy might have been avoidable, 6 (67%) patients underwent the creation of a permanent colostomy. When performing radical surgery with the goal of achieving an R0 resection, it is sometimes not possible to successfully preserve the anus. In addition, our study included eight T4 cases and among them direct tumor invasion to either the bladder and/or prostate was diagnosed in 3 cases before performing neoadjuvant therapy. In these cases, computed tomography and/or magnetic resonance imaging before surgery still showed suspected direct invasion due to the unclear boundary with the surrounding organs. Thus, as a result, abdominoperineal resection was performed in all three cases. However, a pathological study demonstrated the depth of tumor invasion in those three cases to be pCR, T2 and T3, respectively. We therefore need to improve the diagnostic accuracy to avoid extended resection of the anus and the adjacent organs and to consider extending the indications for the preservation of the anal function and the adjacent organs on a case-by-case basis in the future.

The present study was associated with some limitations. First, the study was conducted at a single institution with a relatively small number of patients. In addition, we only examined the short-term results. Only one study from Hiratsuka et al (12) reported the long-term outcomes of CRT with S-1 for locally advanced rectal cancer, noting a 5-year disease-free survival and 5-year overall survival of 66.7 and 74.7%, respectively, for Stage II and III rectal cancer. Therefore, further studies will be needed to confirm the efficacy of preoperative CRT with S-1 for locally advanced rectal cancer, including an assessment of the long-term outcomes regarding disease free survival and local recurrence, and sexual and urinary function.

In conclusion, we reported the efficacy of CRT with S-1. This therapy can be reasonably considered for locally advanced rectal cancer due to its acceptable toxicity and relatively high anti-tumor effect.