Preoperative chemoradiotherapy (CRT) for locally advanced rectal cancer (LARC), followed by standardized total mesorectal excision (TME) surgery, results in excellent local control rates, but distant failure compromises patients’ survival [1, 2]. To reduce distant failure risk, clinical trials aim to intensify systemic treatment, at the hazard of increased toxicity and quality of life impairment [3–5]. Such strategy requires the optimization of any local therapy, including radiotherapy (RT), in terms of efficacy and tolerability.

Advanced RT techniques, namely intensity-modulated radiotherapy (IMRT), volumetric modulated arc therapy (VMAT), and proton therapy showed excellent target volume coverage and organs at risk sparing in dosimetric studies [6–8]. To a very limited extent, clinical studies on LARC irradiation reported enhanced tumor response [9] and reduced acute toxicity [9, 10] when IMRT was compared to conventional 3D conformal radiotherapy (3DCRT). A large-scale direct comparison of clinical results after VMAT and 3DCRT has not been reported to date.

Based on promising dosimetric results, VMAT was introduced to our clinic and gradually replaced 3DCRT for LARC since 2009. The purpose of the present single-center study was to compare VMAT-treated patients with 3DCRT-treated patients in terms of tumor response, acute and late toxicity.

The use of IMRT and VMAT in the preoperative CRT of LARC is still rare, but dosimetric studies showed excellent target volume coverage and organs at risk sparing [7, 8, 10, 17]. Our group already demonstrated the superiority of VMAT over 3DCRT in a planning study with 25 patients, using the treatment protocol of the current study [7]. We analysed, whether these dosimetric advantages translate into significant clinical benefits. When directly comparing VMAT-treated patients with 3DCRT-treated patients in terms of tumor response, acute and late toxicity we found that VMAT provides equal tumor regression combined with reduced acute and late organ toxicity.

To assess tumor response, we used the standardized five-point TRG [14], and found no differences between VMAT-treated patients and 3DCRT-treated patients. Fokas et al. identified TRG as an independent prognostic factor for metastasis-free and disease-free survival of LARC patients treated according to the protocols of the CAO/ARO/AIO rectal cancer trials [18]. Thus TRG may reflect CRT effectiveness and our data could indicate that VMAT-treated patients will experience satisfying oncological outcome. This assumption is supported by a small feasibility study [6], where Richetti et al. analysed VMAT in 45 patients with preoperative CRT of LARC and demonstrated improved dose conformality and a tumor downstaging, comparable to previously published data for conventional RT. Furthermore the favorable oncological outcome in patients treated with IMRT for rectal cancer [9, 19] and other tumors [20–23] supports a potential benefit for VMAT-treated patients. IMRT and VMAT have comparable dosimetric characteristics [17], but IMRT is already used for quite some time and therefore patient data with longer follow up periods are available [24].

In addition, for IMRT, an excellent toxicity profile was achieved for patients with LARC [9, 10, 25], and other tumor entities [20–23, 26]. Comparisons of IMRT with 3DCRT indicate a reduction of acute gastrointestinal toxicity and treatment breaks [10, 25]. Since the issue has not been addressed in larger cohorts, an ongoing clinical trial aims to compare the acute toxicity rates after IMRT and 3DCRT [NCT02151019]. Considerable differences exist between IMRT and VMAT, namely the reduction of treatment delivery time and the number of applied monitor units [6, 27]. Thus, the clinical comparison of VMAT and 3DCRT must be addressed separately. In general, clinical toxicity data for VMAT are scarce. A favorable acute toxicity profile was described for prostate, non-small cell lung, anal canal and endometrial cancer [28–31], and in the feasibility study of Richetti et al. for LARC [6], where high-grade acute organ toxicity occurred in three patients (7 %), only.

The current study demonstrates the potential of VMAT to substantially reduce high-grade acute organ toxicity, showing a ratio of 20 % for 3DCRT versus 5 % for VMAT. Well-known risk factors for the occurrence of high-grade acute organ toxicity in preoperative CRT of LARC are female gender [32], age ≥70 years [33] and low body mass index [34]. In the current study, the VMAT group showed a significantly higher proportion of patients being older than 70 years and having a low body mass index. Remarkably, the VMAT treatment resulted in lower acute toxicity rates despite the preponderance of features for higher risk of toxicity.

With regard to specific organs, differences were found in high-grade skin reaction (7 % vs. 0 %) and in proctitis (12 % vs. 2 %). The reduction of skin reaction might be explained by the fact that VMAT treatment generally leads to an increase in the volume of normal tissue receiving low-dose irradiation and to a decrease in the volume of normal tissue receiving high-dose irradiation [27]. For IMRT in breast cancer treatment, a reduction in acute skin reaction in comparison to 3DCRT was demonstrated [35] where moist desquamation likely occurred in highest skin dose areas [36]. The reduction in proctitis rates was not described previously in the literature. Our group already demonstrated a significant reduction of high dose areas in the PTV with VMAT plans (V_{107 %} = 0.1 %) in comparison to 3DCRT plans (V_{107 %} = 3.5 %) [7]. The diminution of rectal high-dose areas could lead to lower rates of injury to the rectal wall.

In the current study, no differences between 3DCRT treatment and VMAT treatment were observed regarding hematotoxicity. Altogether, six patients (3 %) developed high-grade hematotoxicity. Mell et al. found the irradiated volume of pelvic bone marrow receiving low-dose irradiation to be predictive for hematotoxicity in the CRT of cervical cancer with IMRT. The authors argued that the use of IMRT might be suitable to reduce hematotoxicity in pelvic RT [37]. However, there are no available data comparing hematotoxicity after 3DCRT and VMAT in a similar patient population. Our findings suggest that, despite the probable increase in irradiated volume of bone marrow with VMAT in comparison to 3DCRT, there is no clinically detectable negative effect on hematotoxicity.

To our knowledge, no published data exist on the late toxicity rates after VMAT in preoperative CRT for rectal cancer. We found significantly lower rates of late toxicity for VMAT in comparison to 3DCRT. High-grade late organ toxicity occurred in 22 % of the 3DCRT patients and in 6 % of the VMAT patients. Altogether, high-grade enteritis, urethral stricture or ureteral stenosis occurred in 10 % of the 3DCRT patients, while none of the VMAT patients experienced these complications. In general, the small bowel and bladder complications occur after organ exposure to RT doses of ≥50Gy [38]. Especially for comparably high dose levels, our group demonstrated a remarkable improvement with VMAT. For the small bowel, the V_40Gy was 28.4 % with VMAT plans and 41.8 % with 3DCRT plans. For the urinary bladder, the V_40Gy was 66.5 % with VMAT plans and 88.4 % with 3DCRT plans [7]. These findings could explain the absence of enteritis, urethral stricture and ureteral stenosis in VMAT patients.

To address the limitations of the current study, though patients were treated in accordance with the respective protocol, the comparison of 3DCRT and VMAT was not a predefined endpoint of the trial. Thus, a potential bias due to covariates cannot be excluded with absolute certainty. Furthermore, due to the fact that VMAT was introduced into clinical practice only a short time ago, the groups appear different in length of follow-up. As an important concern for the late toxicity data, a prolonged observation period is required. However, the latency to the occurrence of late toxicity in preoperative CRT of LARC is less well known. After RT of cervical cancer, high rates of urinary tract and small bowel complications were found during earlier follow-up. The complication rates sharply declined after 2–3 years [39]. In the current study, 28 patients (35 %) in the VMAT group were observed for a period of >2 years. Since none of these patients experienced high-grade enteritis, urethral stricture or ureteral stenosis, the current findings indicate that VMAT reduces late toxicity in preoperative CRT of LARC. Nevertheless, the outstanding strength of the current study is the homogeneous treatment according to the German CAO/ARO/AIO-04 trial. The presented data highlight the benefits of the VMAT irradiation for LARC on a powerful basis.

In summary, VMAT treatment in preoperative CRT for LARC showed the potential to substantially reduce high-grade organ toxicity, and lower rates of late toxicity were conceivable. Importantly, we could demonstrate that VMAT irradiation did not impair short-term oncological results. We conclude, that the delivery of preoperative RT using VMAT may pave the way for more efficient systemic therapies, and improved patient survival in the multimodal treatment of LARC.