Abstract
Background
Previous randomized trials found that a prolonged interval between short-course radiotherapy (SCRT, 25 Gy in 5 fractions) and surgery for rectal cancer (4–8 weeks, SCRT-delay) results in a lower postoperative complication rate and a higher pCR rate than SCRT and surgery within a week (SCRT-direct surgery). This study sought to confirm these results in a Dutch national database.
Methods
Patients with intermediate-risk rectal cancer (T3(mesorectal fascia (MRF)–) N0 M0 and T1–3(MRF–) N1 M0) treated with either SCRT-delay (4–12 weeks) or SCRT-direct surgery in 2018–2021 were selected from a Dutch national colorectal cancer database. Confounders were adjusted for using inverse probability of treatment weighting (IPTW). The primary endpoint was the 90-day postoperative complication rate. Secondary endpoints included the pCR rate. Endpoints were compared using log-binomial and Poisson regression.
Results
Some 664 patients were included in the SCRT-direct surgery and 238 in the SCRT-delay group. After IPTW, the 90-day postoperative complication rate was comparable after SCRT-direct surgery and SCRT-delay (40.1 versus 42.3 per cent; risk ratio (RR) 1.1, 95 per cent c.i. 0.9 to 1.3). A pCR occurred more often after SCRT-delay than SCRT-direct surgery (10.7 versus 0.4 per cent; RR 39, 11 to 139).
Conclusion
There was no difference in surgical complication rates between SCRT-delay and SCRT-direct, but SCRT-delay was associated with more patients having a pCR.
Recent randomized trials have shown that a prolonged interval between short-course radiotherapy (SCRT, 25 Gy in 5 fractions) and surgery for rectal cancer (4–8 weeks, SCRT-delay) results in a lower postoperative complication rate and a higher pCR rate than SCRT and surgery within a week (SCRT-direct surgery). This study compared the 90-day postoperative complication rate and pCR rate between SCRT-delay and SCRT-direct surgery using Dutch nationwide real-world data. An advantage in terms of postoperative complications could not be confirmed after SCRT-delay compared with SCRT-direct surgery, but the increased pCR rate after SCRT-delay was confirmed.
Introduction
Total mesorectal excision (TME) preceded by short-course radiotherapy (SCRT, 25 Gy in 5 fractions) has been the recommended treatment strategy for intermediate-risk rectal cancer (T1–3 (without involvement of the mesorectal fascia (MRF–)) N1 M0 and T3cd(MRF–) N0–1 M0) in the Netherlands for over 20 years1. Addition of neoadjuvant SCRT to surgery reduced local recurrence rates in the randomized Swedish rectal cancer and Dutch TME trials2,3. Rather arbitrarily, these trials used a maximum interval of 1 week between completion of SCRT and surgery (SCRT-direct surgery). This short interval remained the standard, backed up by negative results for a slightly longer interval; a retrospective study4 demonstrated an increased risk of postoperative complications when the time between the start of SCRT and TME exceeded 13 days. Subgroup analysis of the Dutch TME trial5 also showed an increased risk of 1-year overall mortality in older patients operated within 4–7 days of completion of SCRT compared with 1–3 days.
More recently, SCRT with a prolonged interval to surgery (4–8 weeks, SCRT-delay) came into focus as a more tolerable neoadjuvant strategy than chemoradiation (50 Gy in 25 fractions combined with a chemosensitizer) for frail patients with locally advanced rectal cancer6,7. The optimal timing of surgery after SCRT was again open to debate. Prospective studies8–10 followed, showing acceptable toxicity and improved tumour downstaging after SCRT-delay. Randomized evidence in favour of SCRT-delay came from the Stockholm III trial11,12, which showed a lower 30-day postoperative complication rate (40.6 versus 52.7 per cent; P = 0.001) and a higher pCR rate (10.4 versus 0.3 per cent; P < 0.001) than after SCRT-direct surgery, at the expense of more acute radiation-induced toxicity grade 3–4 (6.5 versus 0.3 per cent; P < 0.001). The lower postoperative complication rate in the SCRT-delay group was for a large part accounted for by a lower rate of surgical complications (surgical-site infection, deep infection, anastomotic leak, postoperative bleeding, stoma-related complications, wound dehiscence, or other surgical complication).
Since publication of the Stockholm III trial, the Dutch treatment guideline1 has advised discussion of both SCRT-direct surgery and SCRT-delay with patients who have intermediate-risk rectal cancer (shared decision-making). The present study sought to confirm the results of the Stockholm III trial in the Dutch national database using the target trial framework13. The 90-day postoperative complication rate was compared between SCRT-direct surgery and SCRT-delay in an inverse probability of treatment weighted (IPTW) analysis.
Methods
The goal of this study was to evaluate 90-day postoperative complication rates after SCRT-delay versus SCRT-direct surgery for rectal cancer. As the study aimed to answer a causal question using observational data, it was designed according to the target trial framework (Table 1)13.
Table 1.
Summary of protocol of ideal trial evaluating effect of short-course radiotherapy and surgery within a week versus a prolonged interval to surgery on 90-day postoperative complication rate (target trial), emulated using observational data
| Protocol component | Description |
|---|---|
| Eligibility criteria | Patients with intermediate-risk rectal cancer (lower border of tumour below sigmoid take-off and cT3(MRF–) N0 M0 or cT1–3(MRF–) N1 M0 stage) with indication for SCRT and surgery during 2018–2021 in the Netherlands |
| Treatment strategies | A: SCRT followed by TME within 1 week (SCRT-direct surgery) B: SCRT followed by TME within 4–12 weeks (SCRT-delay) |
| Assignment procedures | Random assignment to treatment A or B |
| Follow-up time | Starts at randomization and ends at 90 days after discharge from hospital, death or loss to follow-up |
| Primary endpoint | 90-day postoperative complication rate |
| Secondary endpoints | pCR rate 90-day reintervention Organ failure requiring admission to ICU Postoperative death Anastomotic leakage rate Surgical-site infection rate Abscess not at anastomosis Duration of hospital stay Unplanned readmission within 90 days of initial discharge from hospital |
| Type of analysis | Intention to treat |
MRF, mesorectal fascia; SCRT, short-course radiotherapy; TME, total mesorectal excision.
Patients
Patients who were treated for intermediate-risk rectal cancer (lower border of the tumour below the sigmoid take-off and cT3(MRF–) N0 M0 or cT1–3(MRF–) N1 M0 stage) with SCRT and surgery during 2018–2021 in the Netherlands were included1,14. Exclusion criteria were recurrent rectal cancer, neoadjuvant chemotherapy, resection not by partial mesorectal excision or TME (for example, local excision, sigmoid resection, proctocolecctomy), or TME preceded by local excision. Patients were selected for the SCRT-direct surgery group if the maximum interval between completion of radiotherapy and surgery was 1 week (0–7 days), and for the SCRT-delay group if the interval was 4–12 weeks (28–84 days). This study used a broader interval for SCRT-delay than the Stockholm III trial because some centres in the Netherlands perform response evaluation and surgery at 10–12 weeks after completion of SCRT, in line with the STAR-TREC study15.
Anonymized data were provided by the Dutch Colorectal Audit (DCRA). The DCRA is a mandatory registration that collects patient, tumour, treatment, and surgical and pathological outcome data for all patients who are treated surgically for colorectal cancer in the Netherlands16. Several methods are in place to ensure data validity in the DCRA, as described elsewhere16. Comparison of the DCRA with the Dutch National Cancer Registry showed a case completeness rate of 94 per cent and good correspondence in data (for example, anastomotic leakage rates were 10.4 and 8.7 per cent respectively)17. The clinical audit board of the DCRA approved the research proposal of the present study. No further ethical review was required under Dutch law.
Outcomes
The primary outcome was the 90-day postoperative complication rate, defined as the occurrence of any complication within 90 days after surgery or during the primary hospital stay. Secondary outcomes were the pCR rate (ypT0 N0), 90-day complications requiring reintervention (Clavien–Dindo III), organ failure requiring admission to ICU (Clavien–Dindo IV), death (Clavien–Dindo V), anastomotic leakage, surgical-site infection, abscess not at the anastomosis, duration of hospital stay (number of days between surgery and hospital discharge), and unplanned readmission within 90 days of initial discharge from hospital. Anastomotic leakage was defined by the presence of intra-abdominal fluid or abscess at the anastomosis requiring treatment. This outcome was evaluated only in patients in whom an anastomosis was created.
Statistical analysis
Missing values in baseline patient, tumour, and treatment characteristics were assumed to be missing at random and were imputed using single imputation (Tables S1–S3)18. Confounders were selected based on clinical knowledge. Sex, age, BMI category, history of bowel resection, ostomy before start of treatment, preoperative anaemia, preoperative bowel obstruction, ASA fitness grade, Charlson Co-morbidity Index score, clinical tumour category, clinical nodal status, tumour location, surgical approach, and type of resection were considered to be confounders (Table S2). Confounding was adjusted for using IPTW19. IPTW assigns a weight to each patient, which is calculated as the inverse of the predicted probability of receiving the treatment that was actually received, given the distribution of confounders. Through weighting, a pseudopopulation is created that is well balanced in terms of confounders. This process mimics the exchangeability of groups after random treatment assignment (Table 1). Baseline differences between groups before and after weighting were expressed as the standardized mean difference (SMD), calculated as the mean difference between groups divided by the pooled standard deviation20. An SMD of 0.10 or less was considered to indicated well balanced groups20.
Outcomes were compared between groups using binomial regression for dichotomous outcomes and Poisson regression for count outcomes, both with log link and a robust standard error21,22. P < 0.050 was considered significant. Analyses were repeated with the SCRT-delay group restricted to an interval of 4–8 weeks (28–56 days), so that the present results could be compared directly with those of the Stockholm III trial. Analyses were repeated in the complete-case population to explore the impact of missing data.
The minimal detectable difference was calculated to see whether the sample size was sufficient to confirm the difference in postoperative complication rate (12 per cent absolute difference) that was demonstrated by the Stockholm III trial. Given the present sample size, a postoperative complication rate of 41 per cent, an α of 5 per cent, a power of 80 per cent, and a two-sided alternative hypothesis, this study could detect a difference of 11 per cent or more.
Analyses were done using R version 4.2.0, and packages mice, ipw, survey, sandwich, and EnvStats (R Foundation for Statistical Computing, Vienna, Austria)23–25.
Results
Of 7391 patients in the Netherlands who had surgery for cT1–3 primary rectal cancer during the study interval, 664 were included in the SCRT-direct surgery group and 238 in the SCRT-delay group (Fig. S1).
Before imputation and IPTW, patients in the SCRT-direct surgery group were younger (median 67 (i.q.r. 58–74) versus 68 (60–77) years), had a lower ASA grade (grade I–II: 80.9 versus 72.7 per cent), and more often underwent (L)AR without ostomy (40.6 versus 27.7 per cent). They less often had APR (16.9 versus 21.4 per cent), (L)AR with permanent ostomy (16.4 versus 22.3 per cent) or (L)AR with deviating ostomy (26.1 versus 28.6 per cent) than patients in the SCRT-delay group (Table 2). After imputation and IPTW, baseline characteristics were well balanced.
Table 2.
Patient, tumour, and treatment characteristics of patients with intermediate-risk rectal cancer who had short-course radiotherapy and surgery within a week versus a prolonged interval (4–12 weeks) to surgery, before and after single imputation and inverse probability of treatment weighting
| Before SI and IPTW | After SI and IPTW | ||||||
|---|---|---|---|---|---|---|---|
| SCRT-direct surgery (n = 664) | SCRT-delay (n = 238) | SMD | Missing | SCRT-direct surgery (n = 664) | SCRT-delay (n = 238) | SMD | |
| Sex ratio (F : M) | 234 : 432 | 85 : 153 | 0.010 | 0 (0) | 233.9 : 432.1 | 83.9 : 153.6 | 0.002 |
| Age (years) median (i.q.r.) | 67 (58–74) | 68 (60–77) | 0.175 | 3 (0.3) | 68 (59–75) | 66 (59–75) | 0.001 |
| BMI (kg/m2) | 0.127 | 18 (2.0) | 0.027 | ||||
| Underweight (< 18.5) | 10 (1.5) | 4 (1.7) | 10.5 (1.6) | 3.9 (1.6) | |||
| Normal weight (18.5–24.9) | 237 (36.5) | 100 (42.6) | 254.9 (38.4) | 91.2 (38.4) | |||
| Overweight (25.0–29.9) | 279 (43.0) | 92 (39.1) | 277.6 (41.8) | 96.9 (40.8) | |||
| Obese (≥ 30.0) | 123 (19.0) | 39 (16.6) | 121.0 (18.2) | 45.6 (19.2) | |||
| History of bowel resection | 11 (1.7) | 1 (0.4) | 0.122 | 0 (0) | 8.9 (1.3) | 3.4 (1.4) | 0.008 |
| Ostomy before start of treatment | 11 (1.7) | 6 (2.5) | 0.061 | 5 (0.6) | 11.7 (1.8) | 3.4 (1.5) | 0.025 |
| Preoperative anaemia* | 41 (6.2) | 16 (6.7) | 0.022 | 0 (0) | 42.9 (6.5) | 16.1 (6.8) | 0.013 |
| Preoperative bowel obstruction† | 9 (1.4) | 3 (1.3) | 0.006 | 11 (1.2) | 11.3 (1.7) | 3.6 (1.5) | 0.013 |
| ASA fitness grade | 0.255 | 0 (0) | 0.040 | ||||
| I | 120 (18.1) | 27 (11.3) | 107.6 (16.2) | 35.6 (15.0) | |||
| II | 417 (62.8) | 146 (61.3) | 415.5 (62.6) | 152.9 (64.4) | |||
| III | 123 (18.5) | 61 (25.6) | 135.2 (20.4) | 47.0 (19.8) | |||
| IV | 4 (0.6) | 4 (1.7) | 5.7 (0.9) | 2.1 (0.9) | |||
| CCI score | 0.131 | 219 (24) | 0.022 | ||||
| 0 | 277 (58.7) | 124 (58.8) | 391.0 (58.9) | 140.7 (59.3) | |||
| 1 | 107 (22.7) | 40 (19.0) | 136.0 (20.5) | 48.8 (20.5) | |||
| 2 | 55 (11.7) | 32 (15.2) | 84.1 (12.7) | 30.3 (12.8) | |||
| 3 | 24 (5.1) | 10 (4.7) | 37.6 (5.7) | 12.8 (5.4) | |||
| 4–7 | 9 (1.9) | 5 (2.4) | 15.2 (2.3) | 4.8 (2.0) | |||
| Clinical tumour category | 0.224 | 0 (0) | 0.020 | ||||
| cT1 | 14 (2.1) | 1 (0.4) | 11 (1.7) | 3.6 (1.5) | |||
| cT2 | 135 (20.3) | 40 (16.8) | 129.3 (19.5) | 46.6 (19.6) | |||
| cT3ab | 209 (31.5) | 94 (39.5) | 223.0 (33.6) | 78.4 (33.0) | |||
| cT3x | 148 (22.3) | 48 (20.2) | 144.4 (21.8) | 53.0 (22.3) | |||
| cT3cd | 158 (23.8) | 55 (23.1) | 156.2 (23.5) | 55.9 (23.5) | |||
| Clinical nodal category cN1 | 538 (81.3) | 195 (82.3) | 0.026 | 3 (0.3) | 542.9 (81.8) | 195.4 (82.3) | 0.014 |
| Tumour location‡ | 0.073 | 69 (7.6) | 0.073 | ||||
| Distal (0–3 cm) | 155 (25.4) | 62 (27.8) | 166.6 (25.1) | 59.3 (25.0) | |||
| Midrectal (3–6 cm) | 208 (34.1) | 69 (30.9) | 233.7 (35.2) | 76.3 (32.1) | |||
| Proximal (≥ 6 cm) | 247 (40.5) | 92 (41.3) | 263.7 (39.7) | 101.9 (42.9) | |||
| Surgical approach | 0.115 | 26 (2.9) | 0.044 | ||||
| Laparotomy | 15 (2.3) | 2 (0.9) | 12.5 (1.9) | 4.1 (1.7) | |||
| Laparoscopy | 390 (60.0) | 137 (60.6) | 400.6 (60.3) | 146.6 (62) | |||
| TaTME | 73 (11.2) | 25 (11.1) | 71.9 (10.8) | 22.8 (9.6) | |||
| Robot-assisted laparoscopy | 172 (26.5) | 62 (27.4) | 179.0 (27.0) | 64.0 (27.0) | |||
| Type of resection | 0.289 | 4 (0.4) | 0.030 | ||||
| Extralevator APR | 32 (4.8) | 12 (5.0) | 31.8 (4.8) | 11.4 (4.8) | |||
| Conventional APR | 80 (12.1) | 39 (16.4) | 90.7 (14) | 31.6 (13.3) | |||
| (L)AR with permanent ostomy | 108 (16.4) | 53 (22.3) | 118.2 (17.8) | 41.6 (17.5) | |||
| (L)AR with deviating ostomy | 172 (26.1) | 68 (28.6) | 178.2 (26.8) | 66.8 (28.1) | |||
| (L)AR without ostomy | 268 (40.6) | 66 (27.7) | 245.1 (36.9) | 86.1 (36.3) | |||
Values are n (%) unless otherwise indicated. Differences between groups are expressed as the standardized mean difference (SMD), calculated as the difference between group means divided by the pooled standard deviation. *Haemoglobin below 7 mmol/l in men and below 6.5mmol/l in women. †Admission to hospital or endoscopic intervention for obstructive symptoms. ‡Distance between lower border of tumour and anorectal junction on sagittal MRI. SI, single imputation; IPTW, inverse probability of treatment weighting; SCRT, short-course radiotherapy; CCI, Charlson Co-morbidity Index; TaTME, transanal total mesorectal excision; APR, abdominoperineal resection; (L)AR, (low) anterior resection.
Before IPTW, postoperative complications developed within 90 days in 265 patients (39.9 per cent) in the SCRT-direct surgery group and 101 (42.4 per cent) in the SCRT-delay group (risk ratio (RR) 1.1, 95 per cent c.i. 0.9 to 1.3; P = 0.492) (Table 3). Anastomotic leakage occurred in 71 (16.1 per cent) and 24 (17.9 per cent) respectively (RR 1.1, 0.7 to 1.7; P = 0.627). Similarly, other postoperative complications and duration of hospital stay were comparable between groups. Two patients (0.3 per cent) in the SCRT-direct surgery group and 26 (11.0 per cent) in the SCRT-delay group had a pCR (RR 36, 8.7 to 152; P < 0.001).
Table 3.
Ninety-day postoperative complication and pCR rates after short-course radiotherapy and surgery within a week versus a prolonged interval (4–12 weeks) to surgery, before and after inverse probability of treatment weighting
| SCRT-direct surgery (n = 664) |
SCRT-delay (n = 238) |
Risk ratio* | P | |
|---|---|---|---|---|
| Before IPTW | ||||
| Complication (any) | 265 (39.9) | 101 (42.4) | 1.1 (0.9, 1.3) | 0.492 |
| Anastomotic leakage† | 71 (16.1) | 24 (17.9) | 1.1 (0.7, 1.7) | 0.627 |
| Abscess | 46 (6.9) | 21 (8.8) | 1.3 (0.8, 2.1) | 0.338 |
| Surgical-site infection | 23 (3.5) | 11 (4.6) | 1.3 (0.7, 2.7) | 0.421 |
| Reintervention | 119 (17.9) | 47 (19.7) | 1.1 (0.8, 1.5) | 0.531 |
| Admission to ICU | 54 (8.5) | 14 (6.2) | 0.7 (0.4, 1.3) | 0.285 |
| Death | 4 (0.6) | 4 (1.7) | 2.8 (0.7, 11) | 0.140 |
| Duration of hospital stay (days), median (i.q.r.)‡ | 5 (4–9) | 5 (4–8) | 1.0 (0.8, 1.1) | 0.589 |
| Readmission to hospital | 137 (20.9) | 41 (17.7) | 0.8 (0.6, 1.2) | 0.290 |
| pCR | 2 (0.3) | 26 (11.0) | 36 (8.7, 152) | < 0.001 |
| After IPTW | ||||
| Complication (any) | 266.2 (40.1) | 100.6 (42.3) | 1.1 (0.9, 1.3) | 0.614 |
| Anastomotic leakage† | 69.2 (16.3) | 28.0 (18.3) | 1.1 (0.7, 1.9) | 0.667 |
| Abscess | 46.3 (7.0) | 21.3 (9.0) | 1.3 (0.7, 2.4) | 0.419 |
| Surgical-site infection | 23.3 (3.5) | 10.7 (4.5) | 1.3 (0.6, 2.7) | 0.499 |
| Reintervention | 118.9 (17.9) | 50.7 (21.3) | 1.2 (0.8, 1.7) | 0.345 |
| Admssion to ICU | 56.1 (8.8) | 12.0 (5.4) | 0.6 (0.4, 1.1) | 0.093 |
| Death | 4.0 (0.6) | 3.1 (1.3) | 2.2 (0.6, 7.9) | 0.229 |
| Duration of hospital stay (days), median (i.q.r.)‡ | 5 (4–9) | 5 (4–8) | 1.0 (0.1, 7.1) | 0.974 |
| Readmission to hospital | 138.1 (21.1) | 41.8 (18.0) | 0.9 (0.6, 1.2) | 0.408 |
| pCR | 2.8 (0.4) | 25.5 (10.7) | 39 (11, 139) | < 0.001 |
Outcomes were compared between groups using binomial regression for dichotomous outcomes and Poisson regression for count outcomes, both with log link and a robust standard error. Values are n (%) unless otherwise indicated; *values in parentheses are 95% confidence intervals. *Anastomotic leakage was evaluated only among 440 patients in the short-course radiotherapy (SCRT)-direct surgery and 134 in the SCRT-delay group in the unweighted population, corresponding to 423 and 153 patients respectively in the weighted population) in whom an anastomosis was created. ‡Calculated as number of days between surgery and day of discharge. IPTW, inverse probability of treatment weighting.
After IPTW, 90-day postoperative complications were registered in 266.2 patients (40.1 per cent) in the SCRT-direct surgery group and 100.6 (42.3 per cent) in the SCRT-delay group (RR 1.1, 0.9 to 1.3; P = 0.614). Anastomotic leakage developed in 69.2 (16.3 per cent) and 28.0 (18.3 per cent) respectively (RR 1.1, 0.7 to 1.9; P = 0.667). Other postoperative outcomes remained similar between groups. A pCR occurred in 2.8 patients (0.4 per cent) in the SCRT-direct surgery group and 25.5 (10.7 per cent) in the SCRT-delay group (RR 39, 11 to 139; P < 0.001).
Sensitivity analyses with the interval between completion of radiotherapy and surgery in the SCRT-delay group restricted to 4–8 weeks and complete-case analysis showed similar results (Tables S4–S8).
Discussion
In this study using Dutch nationwide real-world data, the 90-day postoperative complication rate was similar after SCRT-direct surgery and SCRT-delay. The pCR rate was significantly higher in the SCRT-delay group.
This study did not confirm the 12 per cent decrease in postoperative complication rate after SCRT-delay compared with SCRT-direct surgery that was demonstrated in the Stockholm III trial, despite a sufficient sample size. This result was consistent when SCRT-delay was restricted to a 4–8-week interval. The difference between the present results and those of the Stockholm III trial might be explained by improvements in radiotherapy technique since the start of the Stockholm III trial. The Stockholm III trial recruited patients between 1998 and 201311. During the largest part of the study, radiotherapy was administered with a three- or four-beam box technique26. Nowadays, intensity modulated radiotherapy is the standard of care, which has better precision and results in a lower dose to healthy tissues than the three- or four-beam box technique27–29. The authors believe that contemporary radiotherapy techniques increase the risk of postoperative complications to a lesser extent than the technique used in the Stockholm III trial, diminishing the effect of the interval between SCRT and TME on the postoperative complication rate.
The increased probability of a pCR after a prolonged interval between SCRT and TME has been reported consistently in literature30,31. Similarly, a prolonged interval between chemoradiation (50 Gy in 25 fractions combined with a chemosensitizer) and TME for rectal cancer is associated with an improved pCR rate32. Patients who showed a pCR could in theory have been managed by watch-and-wait strategy instead of TME33. This strategy avoids the morbidity of surgery, and has been associated with improved quality of life and less bowel, urinary, and sexual dysfunction34–36. To evaluate eligibility for watch and wait, a response evaluation in patients treated with SCRT-delay has been proposed. The appropriate timing, sensitivity/specificity of the response evaluation, and oncological safety of the watch-and-wait strategy after SCRT are a focus for future research. SCRT-delay and a response evaluation should be offered to patients who are interested in watch and wait.
A substantial proportion of patients with rectal cancer are not interested in a watch-and-wait strategy37,38. Based on the present data, prolonging the interval to surgery does not confer any advantages in terms of postoperative complications in this group. In the Stockholm III trial11, 6.5 per cent of patients in the SCRT-delay group were admitted to the hospital owing to acute radiation-induced toxicity. Again, it is likely that this number overestimates the toxicity rate of current clinical practice because of the older radiotherapy techniques used in the Stockholm III trial. A recent prospective cohort study39 showed no unplanned hospital admissions resulting from radiation-induced toxicity after SCRT-delay, but one in three patients experienced temporary grade III ( severe and disabling but not life-threatening) acute radiation-induced toxicity during the interval. Therefore, SCRT-direct surgery still seems a good option for patients who prefer surgical management.
There are several explanations for the relationship between timing of surgery and the risk of postoperative complications. First, inflammation of the irradiated tissues might impair surgery. Radiation-induced toxicity peaks during weeks 1 and 2 after completion of SCRT and gradually recovers thereafter39. This peak may reflect the least favourable time frame for performing surgery, which is in line with older studies4,5 that showed increased morbidity when SCRT-direct surgery was slightly delayed. In addition, radiotherapy is known to trigger the immune system at a systemic level40. Some studies41–43 have suggested that preoperative radiotherapy impairs the immune response to surgery, which could be measured by a decreased postoperative leucocyte count or a decreased postoperative-to-preoperative leucocyte ratio. The SCRT-delay group in the Stockholm III trial had a significantly higher postoperative-to-preoperative leucocyte ratio than the SCRT-direct surgery group, implying that the immune response had recovered by 4–8 weeks after SCRT43. Another theory is that a prolonged interval increases the risk of pelvic fibrosis. In a non-randomized non-blinded trial44, surgeons scored a higher level of fibrosis in the group that had an 11-week compared with a 6-week interval between chemoradiotherapy and TME. However, this difference did not translate into an increased postoperative complication rate. Finally, a prolonged interval offers the opportunity to improve patient fitness and nutritional status before surgery. Such prehabilitation programmes reduce duration of hospital stay and postoperative complication rates45,46. SCRT-delay combined with a prehabilitation programme might be a good strategy for frail patients.
This study comes with limitations. In some centres in the Netherlands, it is already standard of care to schedule a response evaluation after SCRT-delay and then offer a watch-and-wait strategy in patients with a cCR. Patients managed according to a watch-and-wait strategy were not registered in the DCRA during the study period. The pCR rate in this study will therefore be an underestimation of the real organ-preserving potential of SCRT-delay. The DCRA does not include an explicit definition of postoperative complications. Generally, postoperative complications are defined as ‘any deviation from the normal postoperative course’47. Differences in the interpretation of this definition may have affected the postoperative complication rate in the present study. Despite the data validation mechanisms that ensure the validity of the DCRA, registry studies like this remain prone to misclassification16,17,48,49. This limitation may have led to some effect dilution towards the null. There will always be some residual confounding in non-randomized studies. Because traditionally SCRT-delay was offered to frail patients, residual confounding in the present study will probably disfavour the postoperative complication rate in the SCRT-delay group.
Outcomes after rectal cancer surgery differ between hospitals in the Netherlands, and preferences for SCRT-delay or SCRT-direct surgery probably also differ between hospitals50. Hence, the hospital of treatment should be considered a confounding factor. This was not corrected for here as it did not seem feasible to combine a mixed-effects model with imputation and IPTW.
This study had a sufficient sample size to detect a difference of 11 per cent in the 90-day postoperative complication rate between groups. There was no indication of a difference (there was a non-significant absolute risk reduction of 2 per cent in favour of the SCRT-direct surgery group). Nonetheless, it still is possible that the interval between SCRT and TME had a modest effect on the postoperative complication rate that was not detected.
Supplementary Material
Contributor Information
Maaike E Verweij, Division of Imaging and Oncology, University Medical Centre Utrecht, Utrecht, the Netherlands.
Jolien Franzen, Division of Imaging and Oncology, University Medical Centre Utrecht, Utrecht, the Netherlands.
Wilhelmina M U van Grevenstein, Division of Imaging and Oncology, University Medical Centre Utrecht, Utrecht, the Netherlands.
Helena M Verkooijen, Division of Imaging and Oncology, University Medical Centre Utrecht, Utrecht, the Netherlands.
Martijn P W Intven, Division of Imaging and Oncology, University Medical Centre Utrecht, Utrecht, the Netherlands.
Funding
The authors have no funding to declare.
Author contributions
Maaike Verweij (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Validation, Visualization, Writing—original draft), Jolien Franzen (Data curation, Formal analysis, Investigation, Methodology), Wilhelma Van Grevenstein (Supervision, Writing—review & editing), Helena Verkooijen (Conceptualization, Methodology, Resources, Supervision, Writing—review & editing), and Martijn Intven (Conceptualization, Methodology, Resources, Supervision, Writing—review & editing).
Disclosure
Outside of the submitted work, H.M.V. is a member of the European Commission and the Netherlands Organization of Health Research and Development, and reports grants from Elekta, Sweden, and the Dutch Cancer Foundation. M.P.W.I. has received personal fees from Elekta, Sweden. The authors declare no other conflict of interest.
Supplementary material
Supplementary material is available at BJS online.
Data availability
Raw data were generated at the Dutch Colorectal Audit. Derived data supporting the findings of this study are available from the corresponding author on request.
References
- 1. Richtlijn—Richtlijnendatabase. Primaire behandeling rectumcarcinoom.https://richtlijnendatabase.nl/richtlijn/colorectaal_carcinoom_crc/primaire_behandeling_rectumcarcinoom_bij_crc.html (accessed 24 March 2022)
- 2. Wedish S, Ectal R, Ancer C, Rial T. Improved survival with preoperative radiotherapy in resectable rectal cancer. N Engl J Med 1997;336:980–987 [DOI] [PubMed] [Google Scholar]
- 3. Kapiteijn E, Marijnen CAM, Nagtegaal ID, Putter H, Steup WH, Wiggers Tet al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001;345:638–646 [DOI] [PubMed] [Google Scholar]
- 4. Hartley A, Giridharan S, Gray L, Billingham L, Ismail T, Geh JI. Retrospective study of acute toxicity following short-course preoperative radiotherapy. Br J Surg 2002;89:889–895 [DOI] [PubMed] [Google Scholar]
- 5. Van Den Broek CBM, Vermeer TA, Bastiaannet E, Rutten HJT, Van De Velde CJH, Marijnen CAM. Impact of the interval between short-course radiotherapy and surgery on outcomes of rectal cancer patients. Eur J Cancer 2013;49:3131–3139 [DOI] [PubMed] [Google Scholar]
- 6. Radu C, Berglund Å, Påhlman L, Glimelius B. Short-course preoperative radiotherapy with delayed surgery in rectal cancer—a retrospective study. Radiother Oncol 2008;87:343–349 [DOI] [PubMed] [Google Scholar]
- 7. Hatfield P, Hingorani M, Radhakrishna G, Cooper R, Melcher A, Crellin Aet al. Short-course radiotherapy, with elective delay prior to surgery, in patients with unresectable rectal cancer who have poor performance status or significant co-morbidity. Radiother Oncol 2009;92:210–214 [DOI] [PubMed] [Google Scholar]
- 8. Faria S, Kopek N, Hijal T, Liberman S, Charlebois P, Stein Bet al. Phase II trial of short-course radiotherapy followed by delayed surgery for locoregionally advanced rectal cancer. Color Dis 2014;16:O66–O70 [DOI] [PubMed] [Google Scholar]
- 9. Beppu N, Matsubara N, Noda M, Yamano T, Kakuno A, Doi Het al. Short-course radiotherapy with delayed surgery versus conventional chemoradiotherapy: a comparison of the short- and long-term outcomes in patients with T3 rectal cancer. Surgery 2015;158:225–235 [DOI] [PubMed] [Google Scholar]
- 10. Pach R, Kulig J, Richter P, Gach T, Szura M, Kowalska T. Randomized clinical trial on preoperative radiotherapy 25 Gy in rectal cancer-treatment results at 5-year follow-up. Langenbecks Arch Surg 2012;397:801–807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Erlandsson J, Holm T, Pettersson D, Berglund Å, Cedermark B, Radu Cet al. Optimal fractionation of preoperative radiotherapy and timing to surgery for rectal cancer (Stockholm III): a multicentre, randomised, non-blinded, phase 3, non-inferiority trial. Lancet Oncol 2017;18:336–346 [DOI] [PubMed] [Google Scholar]
- 12. Erlandsson J, Lörinc E, Ahlberg M, Pettersson D, Holm T, Glimelius Bet al. Tumour regression after radiotherapy for rectal cancer—results from the randomised Stockholm III trial. Radiother Oncol 2019;135:178–186 [DOI] [PubMed] [Google Scholar]
- 13. Hernán MA, Robins JM. Using big data to emulate a target trial when a randomized trial is not available. Am J Epidemiol 2016;183:758–764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. D’Souza N, de Neree tot Babberich MPM, D’Hoore A, Tiret E, Xynos E, Beets-Tan RGHet al. Definition of the rectum: an international, expert-based Delphi consensus. Ann Surg 2019;270:955–959 [DOI] [PubMed] [Google Scholar]
- 15. Rombouts AJM, Al-Najami I, Abbott NL, Appelt A, Baatrup G, Bach Set al. Can we Save the rectum by watchful waiting or Trans Anal microsurgery following (chemo) Radiotherapy versus Total mesorectal excision for early REctal Cancer (STAR-TREC study)?: protocol for a multicentre, randomised feasibility study. BMJ Open 2017;7:e019474 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Van Leersum NJ, Snijders HS, Henneman D, Kolfschoten NE, Gooiker GA, ten Berge MGet al. The Dutch surgical colorectal audit. Eur J Surg Oncol 2013;39:1063–1070 [DOI] [PubMed] [Google Scholar]
- 17. DICA . Jaarrapportage 2011. https://dica.nl/jaarrapportage-2013/archief/DICA%20Jaarrapportage%202011.pdf
- 18. Sterne JAC, White IR, Carlin JB, Spratt M, Royston P, Kenward MGet al. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ 2009;339:157–160 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Austin PC. The performance of different propensity-score methods for estimating differences in proportions (risk differences or absolute risk reductions) in observational studies. Stat Med 2010;29:2137–2148 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Austin PC. Using the standardized difference to compare the prevalence of a binary variable between two groups in observational research. Communications in Statistics—Simulation and Computation2009;38:1228–1234 [Google Scholar]
- 21. Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med 2015;34:3661–3679 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Knol MJ, Le Cessie S, Algra A, Vandenbroucke JP, Groenwold RHH. Overestimation of risk ratios by odds ratios in trials and cohort studies: alternatives to logistic regression. CMAJ 2012;184:895–899 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. van Buuren S, Groothuis-Oudshoorn K. Mice: multivariate imputation by chained equations in R. J Stat Softw 2011;45:1–67 [Google Scholar]
- 24. van der Wal WM, Geskus RB. Ipw: an R package for inverse probability weighting. J Stat Softw 2011;43:2–23 [Google Scholar]
- 25. Zeileis A, Köll S, Graham N. Various versatile variances: an object-oriented implementation of clustered covariances in R. J Stat Softw 2020;95:1–36 [Google Scholar]
- 26. Pettersson D, Cedermark B, Holm T, Radu C, Påhlman L, Glimelius Bet al. Interim analysis of the Stockholm III trial of preoperative radiotherapy regimens for rectal cancer. Br J Surg 2010;97:580–587 [DOI] [PubMed] [Google Scholar]
- 27. Urbano MTG, Henrys AJ, Adams EJ, Adams EJ, Norman AR, Bedford JLet al. Intensity-modulated radiotherapy in patients with locally advanced rectal cancer reduces volume of bowel treated to high dose levels. Int J Radiat Oncol Biol Phys 2006;65:907–916 [DOI] [PubMed] [Google Scholar]
- 28. Arbea L, Ramos LI, Martínez-Monge R, Moreno M, Aristu J. Intensity-modulated radiation therapy (IMRT) versus 3D conformal radiotherapy (3DCRT) in locally advanced rectal cancer (LARC): dosimetric comparison and clinical implications. Radiat Oncol 2010;5:17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Wee CW, Kang HC, Wu HG, Chie EK, Choi N, Park JMet al. Intensity-modulated radiotherapy versus three-dimensional conformal radiotherapy in rectal cancer treated with neoadjuvant concurrent chemoradiation: a meta-analysis and pooled-analysis of acute toxicity. Jpn J Clin Oncol 2018;48:458–466 [DOI] [PubMed] [Google Scholar]
- 30. Bujko K, Partycki M, Pietrzak L. Neoadjuvant radiotherapy (5×5 Gy): immediate versus delayed surgery. Recent Results Cancer Res 2014;203:171–187 [DOI] [PubMed] [Google Scholar]
- 31. Wu H, Fang C, Huang L, Fan C, Wang C, Yang Let al. Short-course radiotherapy with immediate or delayed surgery in rectal cancer: a meta-analysis. Int J Surg 2018;56:195–202 [DOI] [PubMed] [Google Scholar]
- 32. Ryan J, O’Sullivan DP, Kelly ME, Syed AZ, Neary PC, O’Connell PRet al. Meta-analysis of the effect of extending the interval after long-course chemoradiotherapy before surgery in locally advanced rectal cancer. Br J Surg 2019;106:1298–1310 [DOI] [PubMed] [Google Scholar]
- 33. Habr-Gama A, Perez RO, Nadalin W, Sabbaga J, Ribeiro U, Silva e Sousa AHet al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. Ann Surg 2004;240:711–718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Hupkens BJP, Martens MH, Stoot JH, Berbee M, Melenhorst J, Beets-Tan RGet al. Quality of life in rectal cancer patients after chemoradiation. Dis Colon Rectum 2017;60:1032–1040 [DOI] [PubMed] [Google Scholar]
- 35. Quezada-Diaz FF, Smith JJ, Jimenez-Rodriguez RM, Nash GM, Guillem JG, Weiser MRet al. Patient-reported bowel function in patients with rectal cancer managed by a watch-and-wait strategy after neoadjuvant therapy: a case–control study. Dis Colon Rectum 2020;63:897–902 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Bach SP, Gilbert A, Brock K, Korsgen S, Geh I, Hill Jet al. Radical surgery versus organ preservation via short-course radiotherapy followed by transanal endoscopic microsurgery for early-stage rectal cancer (TREC): a randomised, open-label feasibility study. Lancet Gastroenterol Hepatol 2021;6:92–105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Couwenberg AM, Intven MPW, Burbach JPM, Emaus MJ, Van Grevenstein WMU, Verkooijen HM. Utility scores and preferences for surgical and organ-sparing approaches for treatment of intermediate and high-risk rectal cancer. Dis Colon Rectum 2018;61:911–919 [DOI] [PubMed] [Google Scholar]
- 38. Gani C, Gani N, Zschaeck S, Eberle F, Schaeffeler N, Hehr Tet al. Organ preservation in rectal cancer: the patients’ perspective. Front Oncol 2019;9:318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Verweij ME, Hoendervangers S, von Hebel CM, Pronk A, Schiphorst AHW, Consten ECJet al. Patient- and physician-reported radiation-induced toxicity of short-course radiotherapy with a prolonged interval to surgery for rectal cancer. Colorectal Dis 2023;25:24–30 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Formenti SC, Demaria S. Systemic effects of local radiotherapy. Lancet Oncol 2009;10:718–726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Fokstuen T, Holm T, Glimelius B. Postoperative morbidity and mortality in relation to leukocyte counts and time to surgery after short-course preoperative radiotherapy for rectal cancer. Radiother Oncol 2009;93:293–297 [DOI] [PubMed] [Google Scholar]
- 42. Hartley A, Giridharan S, Srihari N, McConkey C, Geh JI. Impaired postoperative neutrophil leucocytosis and acute complications following short course preoperative radiotherapy for operable rectal cancer. Eur J Surg Oncol 2003;29:155–157 [DOI] [PubMed] [Google Scholar]
- 43. Pettersson D, Glimelius B, Iversen H, Johansson H, Holm T, Martling A. Impaired postoperative leucocyte counts after preoperative radiotherapy for rectal cancer in the Stockholm III trial. Br J Surg 2013;100:969–975 [DOI] [PubMed] [Google Scholar]
- 44. Garcia-Aguilar J, Smith DD, Avila K, Bergsland EK, Chu P, Krieg RM. Optimal timing of surgery after chemoradiation for advanced rectal cancer: preliminary results of a multicenter, nonrandomized phase II prospective trial. Ann Surg 2011;254:97–102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Gillis C, Buhler K, Bresee L, Carli F, Gramlich L, Culos-Reed Net al. Effects of nutritional prehabilitation, with and without exercise, on outcomes of patients who undergo colorectal surgery: a systematic review and meta-analysis. Gastroenterology 2018;155:391–410.e4 [DOI] [PubMed] [Google Scholar]
- 46. Daniels SL, Lee MJ, George J, Kerr K, Moug S, Wilson TRet al. Prehabilitation in elective abdominal cancer surgery in older patients: systematic review and meta-analysis. BJS Open 2020;4:1022–1041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205–213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Rutegård M, Kverneng Hultberg D, Angenete E, Lydrup ML. Substantial underreporting of anastomotic leakage after anterior resection for rectal cancer in the Swedish Colorectal Cancer Registry. Acta Oncol 2017;56:1741–1745 [DOI] [PubMed] [Google Scholar]
- 49. Gunnarsson U, Seligsohn E, Jestin P, Påhlman L. Registration and validity of surgical complications in colorectal cancer surgery. Br J Surg 2003;90:454–459 [DOI] [PubMed] [Google Scholar]
- 50. Henneman D, Snijders HS, Fiocco M, van Leersum NJ, Kolfschoten NE, Wiggers Tet al. Hospital variation in failure to rescue after colorectal cancer surgery: results of the Dutch Surgical Colorectal Audit. Ann Surg Oncol 2013;20:2117–2123 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Raw data were generated at the Dutch Colorectal Audit. Derived data supporting the findings of this study are available from the corresponding author on request.