In this study, the rates of T-downstaging, N-downstaging and overall downstaging were 36.8%, 32.9%, and 44.7%, respectively. One of three patients with synchronous double primary cancer achieved overall downstaging of primary rectal cancer. The pCR rate was 11.8%, and the overall rate of sphincter preservation was 89.5%. These results were comparable to those of the historical LCRT studies and similar to the two-week course KROG study. The comparable outcomes of our study could be attributed to dose prescription of 33 Gy (BED10, 43.9 Gy), greater than SCRT (BED10, 37.5 Gy) or 35 Gy (BED10, 47.3 Gy), slightly closer to the LCRT (BED10, 59.5 Gy). Oral capecitabine was prescribed as the concurrent chemotherapy agent, followed by delayed surgery with a median interval of 52 days to allow tumor regression before surgery. This study showed comparable toxicities to previous LCRT results. Only four patients (5.3%) developed acute or late grade 3 toxicities and there were no grade 4 toxicities in any patients, including the SIB IMRT 35 Gy patient cohort. One patient received 35 Gy and treated with Taxol and Cisplatin concurrently due to synchronous lung cancer and experienced acute grade 2 diarrhea. All patients had one or more weekend rest without receiving HPCRT and received IMRT to limit the radiation dose to normal organs. We performed SIB IMRT of 35 Gy to the gross tumor to improve the pathologic regression. However, in groups of patients with 35 Gy versus 33 Gy, pCR rate was 12.8% vs. 10.8%, sphincter preservation rate was in 87.2% vs. 91.9%, and 5-year DFS was 85.7% vs. 63.9% (p = 0.067), respectively. None of these differences were statistically significant between the two groups.
Five-year LRFS, DFS, and OS were 90.6%, 73.6%, and 90.6%, respectively, in all patients. These survival rates were comparable to previous LCRT results despite our study including patients with synchronous double primary or distant metastasis. We also included five patients with early clinical stages such as T1-2N0. Of these patients, two had synchronous double primary cancers, two preferred HPCRT followed by local excision rather than upfront radical surgery, and one strongly demanded SCRT due to having to travel long-distances to receive CRT. Of these, four patients underwent sphincter preservation surgery. Thus, all our patients had a median radiotherapy duration of 2 weeks, which is 4 weeks shorter than the conventional 6 weeks of LCRT and because of our shorter CRT duration, they could receive surgery 4 weeks earlier than those receiving traditional LCRT.
In the multivariate analysis for LRFS in this study, only pathologic T-stage was an independent prognostic factor. It is well known that patients with sterilized tumors after preoperative CRT have an excellent prognosis with respect to LRFS or DFS [14, 15]. Nine patients who achieved pCR in our study had experienced no recurrence at the last follow-up. In multivariate analysis for DFS, pathologic N stage and LVI were significant prognostic variables. The pathologic N stage is a well-known prognostic factor for survival and is integral to the American Joint Committee on Cancer staging system [16]. The prognosis of patients with remnant lymph node metastasis after CRT is poor, even in cases of complete primary tumor response [17]. Likewise, patients with more advanced pathologic primary tumor (ypT3-4) with pN0 stage showed slightly better recurrence-free survival or OS than did those with ypT0-2 but positive pathologic N [18]. It is well known that the LVI is an independent prognostic factor for survival. Song JH et al. reported that LVI was a significant prognostic factor affecting distant failure-free survival [19]. Saadoun et al. developed a nomogram with eight variables, including pathological stage, LVI, and PNI, which provided individual risk prediction for recurrence [20]. In our study, DFS of patients with both LVI- and PNI- versus that of patients with LVI + or PNI + differed significantly, although patients with LVI + or PNI + received postoperative adjuvant chemotherapy (often 5-fluorouracil and oxaliplatin) more frequently than did patients who had LVI- and PNI- (18/20 vs. 34/56, p = 0.023). Therefore, more aggressive adjuvant chemotherapy should be considered to improve DFS for these patients.
Some limitations of this study are as follows. First, our study had a small cohort of patients, albeit with a long study duration. We provided this CRT option to selected patients with a preference for HPCRT or even with synchronous distant disease or double primary cancer. Thus, the accrual rate was inevitably slow. However, the treatment protocol was almost the same throughout the entire study period and treatment consistency was maintained. Second, the patient characteristics in our study were heterogeneous, ranging from the early clinical stage to the advanced clinical stage (IVA) or to synchronous double primary cancer. However, all patients experienced the advantages of HPCRT, which are convenience, low cost, shorter radiotherapy time, and earlier surgical or other radical treatments for distant disease or double primary cancer.
In conclusion, the HPCRT of 33 or 35 Gy in 10 fractions showed comparable results to those of historical conventional LCRT studies. This shorter fractionation scheme may be beneficial without reducing oncologic outcomes for patients with early-stage disease, locally advanced rectal cancer, simultaneous distant metastasis, and other double primary cancer requiring early intervention or for patients who cannot attend the hospital multiple times.