Prospective study on dose escalation using intensity modulated radiotherapy (IMRT) with simultaneous integrated boost (SIB) and concurrent chemotherapy in preoperative locally advanced rectal cancer

Abstract

Background

Neoadjuvant chemoradiotherapy (NACTRT) with 45–50.4 Gy in 25–28 fractions, followed by total mesorectal excision and adjuvant chemotherapy is a standard treatment option for locally advanced rectal cancers (LARC). Literature has showed benefit in terms of tumour downstaging and pathological complete response (pCR) rates with radiotherapy dose escalation. The aim of our prospective single-arm study was to analyse the role of this dose escalation, using intensity modulated radiation therapy (IMRT) with simultaneous integrated boost (SIB) in LARC.

Materials and methods

Fifty-two patients of LARC included in the study were treated with NACTRT. Pelvic nodes and primary disease received 45 Gy/25 fractions along with SIB to the gross disease, to a total dose of 52 Gy, along with concurrent capecitabine. Six weeks post-NACTRT, patients underwent pelvic magnetic resonance imaging (MRI) for response assessment and were sent for surgery six-eight weeks after completion of NACTRT, followed by Adjuvant chemotherapy after surgery. Post-op histopathological report was analysed and patients underwent 3-monthly follow up to further ascertain any local recurrences or distant metastasis, disease free survival (DFS) and dose-escalation related toxicities.

Results

Post-NACTRT response evaluation done with Pelvic MRI suggested a complete response or near complete response in 36.54% patients, while sphincter preservation surgery was feasible in 42.30% patients. R0 resection, T and N downstaging, pCR was observed in 94%, 68%, 36% and 26% patients, respectively. The Grade-3 toxicity rates were acceptable. One-year-DFS was 91.84%.

Conclusions

It may be safely concluded that dose-escalated NACRT delivered using IMRT-SIB technique in LARC patients offers decent rates of pCR with acceptable toxicity rates.

Introduction

The standard treatment of locally advanced rectal cancers (LARC) consists of neoadjuvant chemo- radiotherapy (NACTRT) followed by total mesorectal excision [1, 2]. Neoadjuvant radiotherapy improves local control of LARC (T3–4 or N1–2 disease) and, subsequently, can influence overall survival (OS). The standard neoadjuvant radiotherapy treatment in rectal cancer consists of a total dose of 45–50.4 Gy delivered in 25–28 daily fractions [1, 2]. Although higher doses to tumours enhance the response rate, they may also cause more adverse effects and surgical problems. Relatively newer procedures like intensity-modulated radiotherapy (IMRT), volumetric modulated arc therapy (VMAT) and image-guided radiotherapy (IGRT), and other technological breakthroughs in radiotherapy treatment have improved treatment outcomes and decreased side effects [3–5]. Using these planning techniques it is possible to deliver a higher dose to the target avoiding surrounding tissue, thus improving tumour response rate and disease control with the reduction of acute and late toxicities [6–10]. For these reasons, three dimensional conformal radiation therapy (3DCRT) has lately been nearly abandoned and replaced by new innovative techniques for the treatment of rectal cancer [6–11]. Different data in literature have shown a benefit in terms of tumour down-staging and complete response (CR) rates with radiotherapy dose escalation. However, there are only a handful of studies regarding the dose escalation using the relatively newer techniques like IMRT, VMAT and IGRT for the treatment of LARC. The aim of this study was to analyse the role of neoadjuvant radiotherapy dose escalation for the treatment of LARC, using IMRT with simultaneous integrated boost (SIB) for the treatment of LARC.

Materials and methods

Study design

It was single arm prospective study. Preoperative patients with histologically proven LARC attending the Department of Radiation Oncology, Medical College and Hospital, Kolkata, during the study period from September 2022 to February 2024. Patients having either metastasis, Karnofsky Performance Status (KPS) > 60, age > 70 years or prior chemotherapy, prior pelvic radiotherapy or prior pelvic surgery were excluded from the study.

Data collection

Detailed History and Clinical examination findings for each patient was noted at the beginning of the study. Baseline investigations included blood investigations — complete blood count (CBC), kidney function test (KFT), liver function test (LFT), serum carcinoembryonic antigen (S. CEA), imaging investigations — colonoscopy and magnetic resonance imaging (MRI) pelvis, and the preoperative histopathological report.

Prior to treatment initiation, size of the tumour, it’s distance from the anal verge, circumferential resection margin (CRM) and the nodal status were assessed by colonoscopy, MRI pelvis and recorded.

Treatment planning

Positioning and immobilisation

To acquire the planning computed tomography (CT) scan, all patients were immobilised supine with 4 clamp thermoplastic mould with adequate leg rest, using empty bladder protocol (the patient was asked to pass urine just prior to the planning CT scan).

CT simulation

Computed tomography simulation was done using Philips Brilliance 16 slice CT machine and Philips AcQSim Virtual Simulation software. Contouring/target volume delineation was done with the help of Somavision software in accordance with established guidelines [12]. The primary gross tumour volume (GTV-P) was defined as all gross disease on physical examination, endoscopy and imaging. The nodal GTV (GTV-N) included all visible perirectal, mesorectal, and involved iliac lymph nodes. Any doubtful lymph node was also included in GTV-N, in the absence of a biopsy. For low-lying rectal tumours inclusion of the inguinal lymph nodes was also considered. The high-risk clinical target volume (CTV-HR or CTV52) included the GTV with a minimum 1.5–2 cm superior and inferior margin, excluding uninvolved bone, muscle, and air. The entire rectum, mesorectum, and presacral space in the transverse plane at these levels were also included in the CTV52. A 1–2 cm margin into adjacent organs (e.g., bladder, prostate, cervix) was added for T4 tumours. The standard-risk CTV (CTV-SR or CTV45) covered the entire CTVHR, mesorectum, and bilateral internal iliac lymph nodes. The bilateral external iliac and obturator nodes were also included for patients with T4 tumours with anterior organ involvement (bladder, cervix, prostate). If the primary tumour extended inferiorly into the anal canal, the bilateral external iliac as well as the inguinal lymph nodes were included in the CTV45. For the planning target volume (PTV), each CTV, i.e. CTV45 and CTV52, was expanded by 0.5–1 cm to create PTV45 and PTV52, respectively. PTV45 received 45 Gy/25 fractions; PTV52 received 52 Gy/25 fractions, over a period of 5 weeks.

Dose constraints were given to the organs at risk (OARs) [13]:

• urinary bladder (V_40Gy < 40%),

• femoral heads (V₄₀ < 40%),

• small bowel (V_45Gy < 195 cc).

The treatment planning system used was Eclipse v15.5.

Along with this, patients received the usual dose of concurrent oral capecitabine (825 mg/m² BD) on the days of radiation, i.e. 5 days a week, throughout the course of radiation.

Radiological response assessment

An MRI pelvis was obtained at 6 weeks after completion of NACTRT for response assessment, prior to surgery. In conjunction with our Radiology Department the tumour response was graded using the magnetic resonance tumour regression grading (mrTRG) system. Subsequently, they were categorised as CR, partial response, stable disease, progressive disease as per Response Evaluation Criteria in Solid Tumors (RECIST) v1.1

Surgery

Patients underwent surgery 6–8 weeks after completion of NACTRT, by either low anterior resection (LAR) (preferred) or abdominoperineal resection, as per feasibility and surgeon’s discretion.

Pathological response in postoperative histopathological examination (HPE) report was assessed post-surgery and patient was followed up at 3-monthly intervals.

Adjuvant chemotherapy

Post surgery patients were offered adjuvant chemotherapy with CapeOx regimen [oxaliplatin 130 mg/m² intravenous (i.v.) D1, capecitabine per os (p.o.) 1000 mg/m² bis in die (BD) on D1–D14; every 3 weeks for 8 cycles or 6 months].

Follow up

Disease response evaluation was done using clinical examination, S. CEA at each 3-monthly follow up. Contrast-enhanced CT (CECT) thorax and abdomino-pelvic CT scans were done if and when deemed necessary. Radiation induced toxicity assessment was done at each visit using Common Terminology Criteria for Adverse Events (CTCAE), version 5.0.

Statistical analysis

For statistical analysis data was entered into a Microsoft Excel spreadsheet and then analysed by SPSS (Statistical Packaging for the Social Sciences, version 23.0). The results of our study have been denoted using descriptive statistics. The Kaplan–Meier survival analysis was a non-parametric statistic used to estimate the survival function from time data.

Results

Our single institutional study included 52 patients of locally advanced rectal adenocarcinoma. 73.08% were above the age of 50 years and the male:female ratio was 1.36. Patients having preoperative clinical stagings cT3N0, cT4N0 and cT1-4N1-2 accounted for 25%, 21.15%, 53.85%, respectively. Baseline S. CEA was elevated in 34.62% of the population and the CRM was threatened in 61.54 % patients on the pre-op MRI. The patient and tumour characteristics are summarised in Table 1.

Table 1.

Demographic and tumour data

Patient and tumour characteristics		No. of patients (n = 52)
Age		57 (29–69)
Median (range)
Sex
Male		30 (57.69%)
Female		22 (42.31%)
KPS
60		4 (7.69%)
70		10 (19.23%)
80		21 (40.38%)
90		17 (32.69%)
Tumour distance from anal verge [cm]
< 6		15 (28.85%)
6–9		24 (46.15%)
> 9		13 (25%)
cTNM	AJCC 8th stage
cT3N0	II	25%)
cT4N0		11 (21.15%)
cT1–4N+	III	28 (53.85%)
CRM status
Yes		32 (61.54%)
No		20 (38.46%)
S. CEA [ng/mL]
< 5		34 (65.38%)
5–15		12 (23.08%)
> 15		6 (11.54%)

CRM — circumferential resection margin; KPS — Karnofsky Performance Status; S. CEA — serum carcinoembryonic antigen

NACTRT

These patients were treated with neoadjuvant chemoradiation with a radiation dose of 45 Gy in 25 fractions to the pelvic nodes and primary disease along with SIB to the gross disease, to a total dose of 52 Gy. The dose escalation was based on various studies which used the same or similar regimens with promising outcomes [8, 9, 14–19]. Compared to the conventional regimen of 50.4 Gy at 1.8 Gy per fraction, the gross disease received a larger dose of 52 Gy and a higher dose per fraction of 2.08 Gy per fraction. The overall treatment time was shortened from the conventional 28 to 25 days as well. This resulted in the patients receiving a biologically equivalent dose (BED) of 73.63 Gy in comparison to the standard dose prescription of 50.4 Gy at 1.8 Gy per fraction having BED of 68.54 Gy. The dosimetric data has been compiled in Table 2.

Table 2.

Dosimetric data of planning target volume (PTV): PTV45, PTV52 and organs at risk (OARs)

Dosimetric parameters	Mean value	Range
PTV45
Volume [cc]	1312.42	466.9–2206.5
Mean dose [Gy]	46.3	44.8–50.2
D98% [%]	91.42	88–94.1
PTV52
Volume [cc]	631.9	298.2–1000.5
Mean dose [Gy]	52.7	31.5–53.2
D98% [%]	97.44	95.6–99
OAR
Bladder
V40Gy [%]	38.16	32.3–42.8
Small bowel
V45Gy [cc]	156.9	107–199
Right femur
Dmax [Gy]	47.61	43.9–51.3
Left femur
Dmax [Gy]	44.94	39.2–49.8

Treatment interruptions

Out of 52 patients, 5 (9.6%) patients had treatment interruptions during NACTRT due to acute toxicities as represented in Table 3. Out of 52 patients, 46 (88.46%) patients received full dose of concurrent oral capecitabine. 6 patients (11.54%) required dose modifications. Among them, in 5 patients it was due to Gr3 toxicities. In these patients, oral capecitabine dose was reduced to 75% on restart of Radiation treatment. 1 patient had moderate renal impairment prior to start of treatment and received 75% of original dose from day 1 of radiation itself.

Table 3.

Treatment interruptions during neoadjuvant chemoradiotherapy (NACTRT)

No of GAP days	Reason of treatment interruption
6	G3: skin toxicity (week 3)
6	G3: diarrhea (week 2)
5	G3: skin toxicity (week 4)
5	G3: diarrhea (week 3)
4	G3: anemia (week 3)

G — grade

Post-NACTRT response evaluation

The median time from the completion of NACRT to a repeat MRI scan was 52 days (range: 40–68 days). Pelvic MRI suggested a CR or near CR (i.e. mrTRG 1 and 2) in 19 (36.54 %) patients. Partial response was seen in 24 (46.15%) patients. The mrTRG grading and RECIST based radiological response data are presented in Table 4 and Figure 1.

Table 4.

Magnetic resonance tumour regression grading (mrTRG) based response after neoadjuvant chemoradiotherapy (NACTRT)

mrTRG	No. of patients (N = 52)	%
1	9	17.31
2	10	19.23
3	13	25
4	11	21.15
5	9	17.31

Figure 1.

Post neoadjuvant chemoradiotherapy (NACTRT) response on pelvic magnetic resonance imaging (MRI); CR — complete response; mrTRG — magnetic resonance tumour regression grading; PD — progressive disease; PR — partial response; SD — stable disease

Surgery

On pre-surgical evaluation two patients we declared medically unfit due to compromised cardiac function. These two patients were excluded from the late toxicity and response evaluation on follow up in our study and their follow up was done separately. Median time from completion of NACTRT to surgery was 69 days (range: 57–78 days).

Among the 50 patients who underwent surgery, sphincter preservation surgery (i.e. LAR) was deemed feasible by the operating surgeon in 42.30% patients. R0 rates were 94%, T downstaging was observed in 68% patients and N downstaging was observed in 36% of the patients who underwent surgery. The pathological CR (pCR) rate was 26%.

The pathological outcomes are presented in Figure 2.

Figure 2.

Post-surgery response assessment from histopathological report; N — nodal; pCR — pathological complete response; T — tumour

As depicted in Figure 3, 4 patients (8%) had surgical complications: perineal abscess in 2 patients, wound infection in 1 patient and anastomotic leakage in 1 patient, which were managed accordingly.

Figure 3.

Type of surgery done and associated post-surgical complications; APR — abdominoperineal resection; LAR — low anterior resection

Figure 4.

Acute toxicity [during neoadjuvant chemoradiotherapy (NACTRT)] and late toxicity (at 6, 9, 12 monthly follow-up); GI — gastrointestinal; GU — genitourinary

Toxicity evaluation and follow up

The Grade 3 acute haematological toxicity, Grade 3 acute gastrointestinal (GI) toxicity, Grade 3 acute skin toxicity rates were 1.92%,3.85% and 3.85%, respectively, whereas Grade 3 late GI toxicity rates in our study were 0%, 6%, 4.1%, 2.2% during follow up at 3, 6, 9, and 12 months, respectively. The toxicity data are represented in Figure 4. During the follow up period 2 patients developed local recurrences, 2 patients developed distant metastasis and 1 patient was lost to follow up. Therefore, 45 out of 49 patients who were able to complete 12 month follow up remained disease-free and the 1-year DFS in our study was 91.84%. The Kaplan–Meier survival curve is shown in Figure 5.

Figure 5.

Kaplan–Meier survival curve at 1 year

Discussion

Among the varied tools used for disease response assessment in rectal cancer patients, MRI plays a crucial role. Owens et al. [16] conducted a retrospective review on 71 rectal cancer patients inclusive of all stages with similar dose prescriptions as our study, where the mrTRG grades suggesting CR or likely CR was noted to be 47.8% compared to the 36.54% seen in our study.

Jang et al. [20] in their study set out to determine the diagnostic accuracy of mrTRG for pCR and its correlation with pathological findings and found mrTRG 1 to have high specificity for pCR and ≤ ypT1, but suboptimal sensitivity. mr-TRG 1 or 2 showed higher sensitivity for pCR and ≤ ypT1, but lower specificity for the same. Kuo et al. [21] found MRI to have a positive-predictive value of 86.4%, a negative-predictive value of 33.3%, and accuracy of 82.5% in predicting pCR after neoadjuvant CRT.

Patients who exhibit a CR to treatment often have an excellent prognosis and the likelihood of higher DFS rates. This successful response also raises the question whether surgical intervention, such as total mesorectal excision, which carries its own risks, including morbidity, possible mortality, and the chance of requiring a permanent colostomy, is necessary for these individuals. Lately, a trend toward less invasive approaches has been recommended for those who demonstrate a significant or full recovery following initial therapy [20]. In their 2006 study, Habr-Gama et al. [22] reported on the outcomes of a prospective study examining a non-operative “wait-and-see” approach in a specifically chosen cohort of patients who showed no signs of disease clinically or radiologically following NACTRT. The results indicated a promising survival rate with 93% OS and 85% DFS at 5 years. Forgoing surgical procedures is however dependent upon accurately distinguishing those patients who have indeed experienced a total response. Typically, this determination is done through physical exams, endoscopic evaluations, and biopsies, although these techniques are not completely reliable. The efficacy of different imaging tools in reevaluating patients post-chemoradiotherapy has been the focus of numerous research studies, which collectively indicate that MRI, endorectal ultrasound, and ¹⁸F-fluorodeoxyglucose positron emission tomography do not reliably pinpoint patients who have completely responded to treatment, with positive predictive values between 17% and 50% [23–28]. However, a study by Kim et al. [28] involving 40 individuals revealed that incorporating DWI alongside conventional MRI markedly enhanced radiologists’ ability to accurately identify patients who had a CR to treatment, in comparison to using standard MRI techniques alone. As a result, it improved the ability of MRI to predict pCR [29, 30].

Another prospective study by Maas et al. [31] also suggests that a wait-and-see policy for rectal cancer patients with CR (cCR) may be more feasible and safe than surgery. Twenty-one patients with cCR were included in the wait-and-see policy group. Mean follow-up was 25 ± 19 months. One patient developed a local recurrence and had surgery as salvage treatment. The other 20 patients were alive without disease, whereas patients with a pCR after surgery had a 2-year disease-free survival and OS of 93% and 91%, respectively [31].

The wait-and-see approach, or non-operative management after NACTRT, may be a viable alternative to surgical resection, along with the lucrative option of sphincter preservation but its success relies on the multidisciplinary team’s diagnostic skills, patient monitoring, prompt intervention, and patient commitment, and further research on radiological response evaluation tools.

In our study, initial assessment by colonoscopy revealed that 28.85% patients had their tumours located < 6 cm from the anal verge, 46.15% had them between 6 to 9 cm from the anal verge while 25% had it beyond 9 cm from the verge. Sphincter preservation has always been viewed as a major quality of life objective for many patients as well as treating oncologists. Supporting this outlook, two phase 3 trials and a metanalysis including those trials demonstrated that preoperative chemoradiation led to conversion of a group of patients initially deemed to require an abdominoperineal resection to low anterior resection [32–34].

In this study, eventually, 6–8 weeks after neoadjuvant chemotherapy, 53.85% patients underwent APR while sphincter preservation surgery was deemed feasible by the operating surgeon in 42.30% patients, who underwent LAR.

The findings from the long term prospective study by Kang et al. [35] suggests that sphincter preserving surgeries could be a promising substitute for abdominoperineal resection in treating lower rectal cancer, improving quality of life, especially in sexual and urinary functions, without increasing the risk of cancer recurrence.

Delishaj et al. [14] conducted a literature review in 2020, focusing on rectal cancer patients who received escalated radiation doses prior to surgery, revealing a main dose of 55 Gy for high-risk areas and 45 Gy for prophylactic volume, yielding a high R0 resection rate of 98.88% (95–100%) across 16 studies. Additionally, sphincter preservation rates averaged 76.03%, consistent with other studies, including a meta-analysis by Hearn et al. [9], which reported a pooled R0 resection estimate of 90.7%.

Surgical complication data of patients treated to > 54 Gy were available in 25 publications in this meta-analysis. The median overall surgical morbidity was 15% (range: 0–100%) whereas in the review by Delishaj et al. [14] fifteen authors also reported surgical complications and anastomotic leakage in their studies. The mean surgical complications rate there was 15.51% with a range from 0% to 41.9%. Out of the 50 patients in our study who underwent surgery, 4 (8%) had surgical complications.

Among studies which used conventional dose regimens, the study by Jalilian et al. [36] had 127 patients with LARC who received neoadjuvant radiation doses of 45 Gy or 50.4 Gy and achieved pCR rate of 14.96%, while a 2021 study by Suwanthanma et al. [37], analysing 85 patients who had received neoadjuvant chemoradiation at a median dose of 50.4 Gy, found a pCR rate of 21.1% and a disease-free survival rate of 69.7%. Yeung et al. [38] reported a pCR rate of 13.8% with 50.4 Gy during neoadjuvant chemoradiation, alongside T downstaging rates of 49.2% and N downstaging rates of 63.1%.

As already demonstrated in the results of our study, on post-op HPE it was noted that 68% patients had tumour downstaging, 36% patients had nodal downstaging, while pCR rates in our patients were 26%. In a similar dose escalation study by Puri et al. [19] published in 2023, tumour and nodal downstaging was achieved in 78% and 84% of patients, respectively, and pCR was achieved in 22.2% patients. With further dose escalations, as seen in the systemic review and metaanalysis by Burbach et al. [8] including fourteen studies (487 patients treated with ≥ 60 Gy), pCR-rates went as high as 44.4%. All these studies as well as ours point toward the higher response rates in terms of tumour downstaging, nodal downstaging and pathological CR further empowering our idea of dose escalation.

With regards to the concern of toxicity from high radiation doses, our study results clearly demonstrate acceptable acute and late toxicities, where Grade 3 acute hematological toxicities were apparent in 1.92% patients, Grade 3 acute GI toxicity was seen in 3.85% and Grade 3 acute skin toxicity was seen in 3.85% patients. The study by Hearn et al. [9] and, Puri et al. [19] showed acute grade 3 toxicity rates of 9.8% and 10%, respectively. Owens et al. [16] who in their publication stated similar radiation dose delivery of 45 Gy to the pelvis and 52 Gy to the gross disease by the SIB technique, published a 4.2% incidence of grade 3 non-haematological toxicity and 1.5% grade 3 haematological toxicity; 4.2% were admitted during their radiotherapy, and one patient died due to a pelvic abscess. The Grade 3 late GI toxicity rates in our study were 0%, 6%, 4.1%, 2.2% during follow up at 3, 6, 9, and 12 months, respectively.

Standardised contouring guidelines followed ensuring uniformity among treated patients, adequate quality assurance, and collaboration with competent radiologists and pathologists in our hospital strengthened our study. Further, only a handful of studies reported the post-NACTRT radiological response rates.

The main limitations of our study were the absence of randomisation and lack of a direct comparison of our dose escalation technique with standard chemoradiation. Moreover, our follow up period was short; therefore, the need for a longer follow up data and larger sample size to estimate late local recurrences and distant failures cannot be denied.

Conclusions

In our study, LARC patients were treated with 45 Gy in 25 fractions to the pelvic nodes and primary disease along with the SIB to gross disease to a total dose of 52 Gy, along with the standard dose of Concurrent Capecitabine, as a part of NACTRT. Post-NACTRT response evaluation done with pelvic MRI suggested a CR or near CR in 36.54% patients, while sphincter preservation surgery was deemed feasible by the operating surgeon in 42.30% patients and the pCR rate was 26%. The toxicities were well within the acceptable range and our patients had a DFS of 91.84% at 1 year. Thus, it may be safely concluded that dose-escalated NACRT therapy with 45 Gy in 25 fractions to the pelvic nodes and primary disease along with SIB to gross disease to a total dose of 52 Gy in LARC patients offers decent rates of pCR with acceptable toxicities.

Data availability statement

The dataset used and analyzed during the study is available from the corresponding author on reasonable request.