Efficacy and safety of short-course radiotherapy versus total neoadjuvant therapy in older rectal cancer patients: a randomised pragmatic trial (SHAPERS)

Abstract

Although total neoadjuvant therapy (TNT) is a new standard of care for locally advanced rectal cancer (LARC), there are no data to confirm the safety and efficacy of this approach in older patients. SHAPERS is a multicentre, open-label, randomised pragmatic trial, aiming to assess whether neoadjuvant short-course radiotherapy (SCRT) is a better trade-off between safety and efficacy than TNT in LARC patients aged ≥70 years. Eligible patients are randomised in a 1 : 1 ratio to SCRT followed by surgery [or watch & wait (w&w)] ± adjuvant chemotherapy or TNT (either SCRT followed by 12-18 weeks of chemotherapy, or long-course chemoradiotherapy followed or preceded by 16 weeks of chemotherapy, based on the investigator’s choice) followed by surgery (or w&w). The primary endpoint is the net treatment benefit, a multicomponent measure of treatment effect based on generalised pairwise comparisons, and defined by four prioritised outcome measures: (i) overall survival at 3 years; (ii) progression-free survival at 3 years; (iii) increased-grade peripheral sensory neuropathy at 3 years; (iv) grade ≥3 toxicities during treatment. The study sample size includes 230 eligible patients, to be recruited at 15-20 centres in Belgium. The trial is registered with ClinicalTrials.gov (NCT06052332).

Graphical abstract

Highlights

• •TNT is a standard treatment for LARC.
• •Safety and efficacy of TNT in older patients are unknown.
• •SHAPERS compares neoadjuvant SCRT and TNT in LARC patients aged ≥70 years.
• •The primary endpoint is the NTB, a multicomponent measure based on GPC.
• •Results of this trial will inform the management of this unserved patient population.

Introduction

Total neoadjuvant therapy (TNT) has recently become a new standard of care for locally advanced rectal cancer (LARC), following the results of two phase III trials.^1,2 In RAPIDO, high-risk LARC patients who had received short-course radiotherapy (SCRT) followed by 18 weeks of oxaliplatin-based chemotherapy had a better 3-year disease-related treatment failure than those treated with long-course chemoradiotherapy (LC-CRT) followed by surgery ± adjuvant chemotherapy [hazard ratio (HR) 0.75, P = 0.019].¹ This benefit did not translate into an overall survival (OS) advantage (HR 0.91, P = 0.50), while TNT-treated patients suffered higher local recurrence rates (11.7% versus 8.1%, P = 0.07).^1,3 In PRODIGE-23, administering mFOLFIRINOX before LC-CRT, surgery, and adjuvant chemotherapy was superior to LC-CRT followed by surgery and adjuvant chemotherapy, in terms of disease-free survival (DFS) (HR 0.69, P = 0.034) and OS (HR 0.73, P = 0.033).^2,4 Other TNT strategies are used in clinical practice. In the non-comparative, randomised phase II OPRA trial, a strategy of ‘watch & wait’ (w&w) following LC-CRT with 16 weeks of either induction (INCT) or consolidation (CNCT) oxaliplatin-based chemotherapy was investigated.⁵ While the sequential order of LC-CRT and systemic chemotherapy did not influence DFS, patients treated with CNCT had higher chances to preserve the rectum (54% versus 39% at 5 years, P = 0.012).⁶ Survival outcomes were similar to those historically reported with LC-CRT alone.

Concerns remain about the routine use of TNT. Adding systemic chemotherapy to SCRT or LC-CRT increases the toxicity burden.^1,2 Furthermore, it is unknown whether the available data are applicable, both in terms of efficacy and toxicity, to an older population. As it often happens in clinical research,⁷ older patients were underrepresented in these trials. In PRODIGE-23, eligibility was restricted to patients ≤75 years old, and only 12% were ≥70 years.² While no age limit was set in RAPIDO, 40% were ≥65 years old, in contrast with real-life data suggesting that 55% of colorectal cancer patients are ≥65 years.^1,8 The median age was 61 years in both trials, 59 years in the OPRA INCT-CRT arm, and 56 years in the OPRA CRT-CNCT arm.⁶ The contention that older rectal cancer (RC) patients represent a unique population, for which results from age-unselected trials may not hold true, is supported by the results of the NACRE trial.⁹ While LC-CRT and SCRT are known to be equivalent with regard to safety and efficacy,^10,11 ≥75-year-old LARC patients treated with LC-CRT in this study suffered more grade ≥3 adverse events (AEs) and grade ≥3 post-operative toxicities than those treated with SCRT. With the caveats of the premature study discontinuation, statistically significant differences in OS (HR 3.6, P = 0.04) and cancer-specific survival (HR 4.9, P = 0.03) favoured the SCRT arm.⁹

The advent of TNT has eventually generated a new unmet need for LARC. Studies are needed to evaluate the overall impact of TNT in older patients as compared with conventional treatments, not only in terms of survival but also in terms of toxicity, long-term functional sequelae, and quality of life (QoL).

Methods

Study design

SHAPERS is a multicentre, open-label, randomised pragmatic trial. The study design reflects current practice in Belgium, and it is shown in Figure 1 and the Graphical Abstract. Eligible LARC patients are randomised in a 1 : 1 ratio to the SCRT or TNT arm. Stratification factors include mesorectal fascia involvement, tumour location, and TNT regimen chosen by the investigator before randomisation.

Figure 1.

Study design.

AE, adverse event; ECOG PS, Eastern Cooperative Oncology Group performance status; OS, overall survival; PFS, progression-free survival; RC, rectal cancer; SCRT, short-course radiotherapy; TNT, total neoadjuvant therapy; w&w, watch & wait.

^aHigh-risk features include: T4, extramural invasion (EMVI+), mesorectal fascia involvement/threatening.

^bAccording to the decision of the investigator.

The SCRT arm consists of SCRT (25 Gy in 5 fractions) followed by surgery [or w&w if clinical complete response (cCR)], and optional adjuvant chemotherapy. Surgery is carried out 6-10 weeks later. The TNT arm includes four regimens, as follows:

• -RAPIDO: SCRT followed by up to 18 weeks of oxaliplatin-based chemotherapy (i.e. nine cycles of FOLFOX or six cycles of CAPOX), and surgery (or w&w if cCR)
• -RAPIDO light: SCRT followed by up to 12 weeks of oxaliplatin-based chemotherapy (i.e. six cycles of FOLFOX or four cycles of CAPOX), and surgery (or w&w if cCR)
• -OPRA INCT-CRT: up to 16 weeks of oxaliplatin-based chemotherapy (i.e. eight cycles of FOLFOX or five cycles of CAPOX), followed by LC-CRT (i.e. 25-28 fractions of 1.8-2.0 Gy each ± a boost to the primary tumour and involved lymph nodes for a total of 50-56 Gy with concurrent capecitabine or continuous infusion 5-fluorouracil) and surgery (or w&w if cCR)
• -OPRA CRT-INCT: LC-CRT followed by up to 16 weeks of oxaliplatin-based chemotherapy (i.e. eight cycles of FOLFOX or five cycles of CAPOX) and surgery (or w&w if cCR)

Imaging assessment

Imaging scans [magnetic resonance imaging (MRI) of the rectum and computed tomography (CT) of the thorax, abdomen, and pelvis, or CT of the thorax and MRI of the abdomen and pelvis] are carried out at baseline, and at the end of the neoadjuvant treatment.

Geriatric assessment and QoL questionnaires

A geriatric evaluation with the G8 Geriatric Assessment Screening Tool is carried out at baseline, after completion of neoadjuvant therapy, and at the end-of-treatment (EOT) visit. Four QoL questionnaires are used: the European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30, EORTC QLQ-CIPN20, EORTC QLQ-CR29, and EQ-5D-5L. These are collected at baseline, after completion of neoadjuvant therapy, at the EOT visit, at each follow-up visit for the first 3 years, and every year thereafter.

Follow-up schedule

Follow-up visits start 3 months after the EOT visit, and continue every 3 months for the first and second year, every 6 months for the third, fourth, and fifth year, and then once at the end of the sixth year. Patients undergo imaging scans every 6 months for 5 years, while a colonoscopy is carried out at the end of the first and fourth years. A more intensive follow-up especially for the first 3 years is adopted for patients who are managed according to w&w. From the sixth-year follow-up visit onwards, survival follow-up for each patient is carried out until all patients have had their sixth-year follow-up visit as applicable.

Study objectives

The primary objective is to assess whether SCRT is a better trade-off between efficacy and safety than TNT in an older RC population. The secondary objectives are to assess short- and long-term efficacy, safety, and QoL of the study treatments, and to evaluate the impact of these in terms of use of health care resources.

Study endpoints

The primary endpoint of the trial consists of four prioritised outcomes:

• 1.OS within 3 years
• 2.Progression-free survival (PFS) within 3 years
• 3.Increased-grade peripheral sensory neuropathy at 3 years
• 4.Grade ≥3 toxicities during treatment

OS is calculated from randomisation to death from any cause. PFS is calculated from randomisation to any of the following events: unresectable tumour due to local tumour progression, R2 resection of the primary tumour, locoregional recurrence after an R0/R1 resection, distant metastases, or death from any cause. Increased-grade peripheral sensory neuropathy is assessed 3 years after randomisation by the investigator according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events (NCI-CTCAE) v5.0, and is defined as any-grade increase as compared with baseline. Grade ≥3 toxicities during treatment (i.e. from the first day of treatment until the EOT visit) are assessed by the investigator according to the NCI-CTCAE v5.0.

The secondary endpoints include:

• -Safety
• -QoL
• -Compliance to treatment
• -Complete response
• -R0 resection
• -Organ preservation
• -Use of health care resources

Statistical analysis

The net treatment benefit (NTB), a multicomponent measure based on generalised pairwise comparisons (GPCs), will be used as the measure of treatment effect to compare the two interventions on the four prioritised outcomes of interest. GPC uses all pairs of patients formed by taking one patient from the SCRT arm and one from the TNT arm.¹² Focusing first on the outcome of highest priority, each pair thus formed yields a pairwise score, which is equal to +1 if the pairwise comparison favours SCRT (‘favourable’), to −1 if it favours TNT (‘unfavourable’), and to 0 (‘neutral’) if the pairwise comparison favours neither SCRT nor TNT (i.e. both patients in the pair have an equivalent outcome for that measure). If the pairwise score is +1 or −1 for the outcome of highest priority, a pair is definitely classified, and the next pair is assessed. If the pairwise score is 0, the next prioritised outcome on a prioritised list is used to classify the pair as favourable, unfavourable, or neutral. The process is repeated until each pair is classified as favourable or unfavourable, or all outcome measures have been used and a pair is still neutral. The calculation of the NTB for the four prioritised outcomes is illustrated in Figure 2. The NTB is estimated by the proportion of favourable pairs minus the proportion of unfavourable pairs. Given that toxicity outcomes are ranked third and fourth, pairwise comparisons for these outcomes are conditional to survival and disease progression times being clinically similar, i.e. not larger than the 12-month threshold of clinical relevance according to the patients’ feedback (see below), thereby ensuring that toxicity gains are not obtained at the price of either decreased survival or disease recurrence. The statistical null and alternative hypotheses are expressed in terms of the NTB at 3 years: H₀: NTB _[3] = 0 versus HA: NTB _[3] ≠ 0. Rejection of the null hypothesis would imply that patients in the SCRT arm have a different overall outcome (taking all four outcome measures into account) than those in the TNT arm.

Figure 2.

Calculation of the net treatment benefit based on the four prioritised outcome measures.

OS, overall survival; PFS, progression-free survival.

Sample-size calculation was carried out via simulations. Five thousand trials were generated with increasing sample sizes, equally divided between the two treatment groups. For each simulated trial, assumptions for the proportions of events for each of the measures composing the NTB were:

• -Death rate at 3 years: 15% for the SCRT arm and 12% for the TNT arm
• -PFS event at 3 years: 32% for the SCRT arm and 24% for the TNT arm
• -Increased-grade neurological toxicity at 3 years: 5% for the SCRT arm and 35% for the TNT arm
• -Grade ≥3 toxicities during treatment: 24% for the SCRT arm and 55% for the TNT arm

Under these assumptions, a sample size of 115 patients per group would lead to a power of 80% with a two-sided type I error of 5% to demonstrate the superiority of the SCRT arm over the TNT arm. Overall, 264 patients are needed to be screened to have 230 assessable patients. For the purpose of this trial, a patient is deemed assessable if he/she is eligible and randomised into the study.

No formal interim analysis is planned for this trial. Nevertheless, an independent data monitoring committee will review safety data and advise on the study continuation, once 40% of the patients have completed the study treatment.

Patients’ involvement in the study design

When designing the study, feedback provided by RC patients and patient representatives from European Cancer Patient Coalition (ECPC), Digestive Cancers Europe (DCE), and Patients de l’Institut Jules Bordet qui Soutiennent et Améliorent la Recherche en Oncologie (PISARO) was taken into account. They ranked the four study outcome measures in a hierarchical list, which reflected their priorities when choosing between different treatment options. Also, they indicated a minimum survival gain of 12 months to accept an increased risk of short- and long-term treatment-related toxicity.

Eligibility criteria

Inclusion and exclusion criteria are reported in Table 1.

Table 1.

Eligibility criteria

Inclusion criteria 1.Age ≥70 years old 2. ECOG performance status (PS): ○ ≤1 if age >75 years old ○ ≤2 if age ≤75 years old 3.Histologically or cytologically confirmed adenocarcinoma of the rectum 4.Distal border of the tumour below the peritoneal reflection and within 15 cm of the anal verge 5.Operable stage III or high-risk stage II RC [high-risk tumours defined as those having ≥1 of the following features: T4, MRF involvement/threatening (tumour within 1 mm of the MRF), EMVI]. Patients with involvement of lateral pelvic lymph nodes are also eligible 6. Adequate bone marrow function as defined below: ○ Absolute neutrophil count ≥2000/μl ○ Haemoglobin ≥9 g/dl ○ Platelets ≥100 000/μl 7. Adequate liver function as defined below: ○ Serum total bilirubin ≤1.5 × ULN. In case of known Gilbert’s syndrome <3 × UNL is allowed ○ AST (SGOT) and ALT (SGPT) ≤2.5 × ULN ○ Alkaline phosphatase ≤2.5 × ULN 8.Adequate renal function as defined by estimated glomerular filtration rate (GFR) ≥30 ml/min/1.73 m2 (according to the CKD-EPI 2021 equation) 9.Absence of clinical conditions that in the opinion of the investigator would contraindicate neoadjuvant therapy and/or surgery 10.Signed informed consent form (ICF) obtained before any study-related procedure 11.Male subjects with partners of childbearing potential must agree to use condom during the course of this study and for at least 6 months after the last administration of study drugs

Exclusion criteria 1.Extensive growth into the cranial part of the sacrum (above S2/3 junction) or the lumbosacral nerve roots indicating that surgery will never be possible even if substantial tumour down-sizing is achieved 2.Presence of metastatic disease or recurrent rectal tumour 3.Presence of grade ≥2 peripheral neuropathy according to the Common Toxicity Criteria for Adverse  Events (CTCAE) v.5.0 4.Significant medical, neuropsychiatric, or surgical condition, currently uncontrolled by treatment, which, in the principal investigator’s opinion, may interfere with completion of the study 5.Any contraindication to pelvic irradiation as evaluated by the investigator 6.Known hypersensitivity reactions to the study drugs or to any excipients, premedications or noninvestigational medicinal products or concomitant medications 7.Any investigational anticancer therapy other than the protocol-specified therapies (participation in other prospective studies which do not imply any specific intervention may be allowed after discussion with the study chair) 8.Patients with a prior or concurrent malignancy whose natural history or treatment has the potential to interfere with the safety or efficacy assessment 9.Clinically significant (i.e. active) cardiovascular disease: cerebral vascular accident/stroke, myocardial infarction, unstable angina, congestive heart failure (grade III or IV as classified by the New York Heart Association), or serious cardiac arrhythmia requiring medication within the past 6 months 10.Known complete DPD deficiency 11.Any previous treatment for RC 12.Use of brivudine, sorivudine, or their chemically related analogues

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CKD-EPI, Chronic Kidney Disease-Epidemiology Collaboration; DPD, dihydropyrimidine dehydrogenase; EMVI, extramural vascular invasion; MRF, mesorectal fascia; RC, rectal cancer; SGOT, serum glutamic oxaloacetic transaminase; SGPT, serum glutamic-pyruvic transaminase; ULN, upper limit of normal.

Sponsorship, ethical approval, and participating centres

The SHAPERS trial is sponsored by the Institut Jules Bordet, endorsed by the Belgian Group of Digestive Oncology (BGDO) and the Oncodistinct Network, and funded by the Belgian Health Care Knowledge Centre (KCE). The study is meant to be open at 15-20 Belgian centres, and accrual is expected to be completed in 36 months. The trial is approved by the Belgian ethics committees and regulatory authorities, and registered with ClinicalTrials.gov (NCT06052332).

Discussion

Conducting ad hoc clinical trials for older cancer patients has been identified as a priority by several academic organisations such as the American Society of Clinical Oncology (ASCO), the Society of Geriatric Oncology Consensus (SGOC), and the Cancer and Aging Research Group (CARG). Also, the International SGOC consensus in older patients with cancer recommends that future efforts should also focus in designing older-patients-oriented trials that integrate geriatric assessment parameters and address key geriatric outcomes, clinical needs, and personal preferences of this population.¹³ Nevertheless, accrual is still challenging for such trials, with a high risk of premature termination or the research question becoming obsolete before the trial is completed.¹⁴ Conventional designs based on non-inferiority considerations might therefore not be advisable in such contexts. Instead, there is a pressing need for innovative designs, such as those explicitly focusing on ensuring sufficient power for a single hypothesis-testing framework with both efficacy and safety outcomes considered at once.¹⁵

The SHAPERS trial is the first study to compare conventional neoadjuvant therapy regimen with TNT in older patients with LARC. By using a novel and patient-centred endpoint, this study has the potential to establish the optimal care in this setting, providing physicians and patients with all-round, comprehensive evidence that can inform the decision making, and improve overall patient experience during treatment, as well as survivorship after treatment.