Preoperative, Adaptive Radiotherapy Concomitant to Chemotherapy for Rectal Adenocarcinoma (Adaptive Rectal Cancer Trial 02)

Background and rational

Preoperative radiochemotherapy (RCT) has a beneficial impact on loco-regional control, improves pathological complete response (pCR) rate and has been considered a standard therapeutic option in advanced, resectable, rectal adenocarcinoma. In the last years, three phase III trials using total neoadjuvant therapy (TNT), that is radiotherapy and all cycles of chemotherapy before surgery, or intensified preoperative chemotherapy have changed this standard.

The RAPIDO trial included MRI-diagnosed LARC patients with either cT4a/b, extramural vascular invasion, cN2, involved mesorectal fascia or enlarged lateral lymph nodes considered to be metastatic. Patients were randomized to short course radiotherapy (SCRT), 5Gyx5 fractions, with subsequent six cycles of CAPOX or nine cycles of FOLFOX4 followed by total mesorectal excision (TME) (experimental arm) or capecitabine-based chemoradiotherapy (25-28 x 2.0-1.8 Gy) followed by TME and optional, predefined by hospital policy, postoperative eight cycles of CAPOX or twelve cycles of FOLFOX4 (standard arm). The main end point was to decrease Disease-related Treatment Failure (DrTF), defined as locoregional failure, distant metastasis, a new primary colorectal tumor or treatment-related death. 920 patients were enrolled. At three years, cumulative probability of DrTF was 23.7% in the experimental arm and 30.4% in the standard arm (HR 0.76 [0.60 - 0.96]; p = 0.02). ypT0N0 rate was 28% in TNT arm and 14% in standard arm (p<0.0001). The 3-years probability of distant metastasis was 20.0% in the experimental group and 26.8% in standard group (p=0.0048). The 3-years probability of locoregional failure was 8.3% and 6.0% in experimental and standard groups, respectively (P=0.12) (1).

The UNICANCER-PRODIGE 23 Trial randomized 461 patients staged cT3 (considered at risk of local recurrence and for which a multidisciplinary board recommended chemoradiotherapy) or cT4 to either neoadjuvant chemotherapy consisting of six cycles of folfirinox, chemoradiotherapy and surgery (experimental arm), or chemoradiotherapy followed by surgery (standard arm). Adjuvant chemotherapy consisted of modified Folfox6 for 6 cycle or 4 cycles of capecitabine months for experimental arm and 12 cycles of modified Folfox6 or 8 cycles of capecitabine for standard arm. Main end point was 3-years DFS. At a median follow up of 46.5 months, 3-years DFS were 76% in experimental arm and 69% in standard arm (p=0.034). ypT0N0 rates was significantly increased in experimental arm, 28% vs 12%, p<0.0001. 3-years metastasis free survival was 79% in the experimental group and 72% in the standard group (p=0.017). No difference in locoregional control was seen in the two groups (4% vs 6%) (2).

The STELLAR Trial enrolled cT3, cT4 or N positive stage and randomized them to short course RT (5x5Gy) followed by 4 cycles of Capox, the TNT arm (n=302 patients) or chemoradiotherapy, the standard arm, (n=297 patients). After total mesorectal excision, TNT arm receive 2 cycles of adjuvant Capox, the standard arm 6 cycles of Capox. Primary end point was 3-years DFS. At a median follow up of 35 months, 3-years DFS were 64.5% and 62.3 in TNT and standard arm, respectively (p<0.01). ypT0N0 or sustained cCR were 21.8 % in TNT arm and 12.3% in standard arm, p=0.002). There was no significant difference in metastasis free survival and locoregional relapse between the two groups (3).

Total neoadjuvant therapy or chemotherapy intensification in preoperative treatment has become a new standard included in international guidelines such as the NCCN guidelines (4) guidelines 2024). The above mentioned trials showed that distant control was still the main problem, however, TNT or induction chemotherapy intensification combined with radiotherapy provided significant improvement in distant disease free survival as well as in the rate of pathological complete remissions, indicating that this is the right path.

A new treatment option is now available for rectal cancer: the watch and wait strategy (WaW). After the pioneering publication by Habr-Gama (5), at least two large multicenter, retrospective, studies (6,7), two meta-analysis (8,9), a randomized phase II trial, the OPRA trial (10), and a phase III trial, the OPERA trial (11), show that watch and wait strategy is feasible without any negative effect on clinical outcome.

The achievement of a high pCR rate is still an important endpoint because it is strongly associated with good prognoses (12) and could maximize the probability of WaW strategy or conservative surgery in low rectal cancer.

In an exploratory study investigators tested feasibility and effectiveness of preoperative adaptive radiotherapy (RT) consisting of delivering a simultaneous integrated boost (SIB) to the residual tumor visible on the images of a simulation CT and MR acquired in the middle of chemoradiation treatment (13).

The rationale of the adaptive RT is based on the following considerations: 1) residual tumor may represent a more resistant component of disease for which it may be appropriate to deliver a higher dose; 2) due to the early reduction of the tumour during RCT, a boost to residual tumor in the final phase of treatment allows a reduction of volumes treated with the higher dose (14); 3) limiting the boost to the final phase of the treatment in "few" fractions is expected to reduce the risk of early treatment interruptions due to toxicity; 4) in a previous study it was shown that rectal mobility is reduced in the second part of the treatment, permitting the application of reasonably small margins (15).

After the early report, the clinical experience with adaptive RT concomitant to cisplatin and fluoropyrimidine was consolidated and a radiobiologic index based on the Early Regression Index_Tumor Control Probability (ERI_TCP) during neo-adjuvant radiochemotherapy was applied to predict the pCR rate (16). In brief, ERI-TCP is a value calculated by the following formula:

ERITCP=-ln[(1-(Vmid/Vpre))

Vpre Vpre is the initial volume of the tumor Vmid is the volume of residual tumor still visible on the images of a MRI performed in the middle of radiotherapy treatment.

This index proved to be a highly reliable predictor of pCR and complete clinical response as showed in the figure 1 and 2. In particular, for ERI_TCP value>32,6, the probability of pCR or cCR is zero.

Subsequently, this index was found to be an excellent predictor of outcome (17) and was used to investigate the impact of change of oxaliplatin (OXA) dose on the pCR and residual vital cell (RVC) fraction rate (18).

The experience with adaptive radiotherapy was more recently updated and the results sent to ESTRO meeting 2024 (19), summarized as follows. Patients with low T2N0 surely eligible for abdominal-perineal resection, T3/T4N0 or any T N+ rectal adenocarcinoma were considered. Concomitant chemotherapy consisted of Oxaliplatin 85 mg/m2 on days -14, 0, +14, and capecitabine 850 mg/m2 BID from day -14 to the end of radiotherapy (day 0 was the start of radiotherapy). From 2nd semester 2015 to December 2018 in the attempt to find a better balance between efficacy and toxicity, our medical oncologist team decided to prescribe only two cycles of oxaliplatin on day -14 and 0. Radiotherapy consisted in 41.4Gy in 18 fractions, 2.3 Gy/fraction, delivered with Tomotherapy or VMAT to the tumor and regional lymph-nodes expanded by 5 mm (PTV1) defined on simulation CT and MRI imaging. After 9 fractions (half treatment) CT and MRI were repeated for the planning of the adaptive phase: PTVadapt was generated by adding a 5mm margin to the residual tumour still visible on half treatment MRI. On the last 6 fractions, a boost of 3.1 Gy/fr delivering a total dose of 46.2 Gy in 18 fractions was delivered to PTVadapt while concomitantly delivering 2.3 Gy/fr to PTV1.

From September 2009 to March 2023, 152 patients (pts) were treated. Toxicity. Seventeen pts (11%) experienced G3 gastrointestinal toxicity: diarrhea 10 (6.5%), proctitis 6 (4%), rectal bleeding 1 (0.5%), nausea/vomiting 1 (0.5%). Five pts had G3 skin toxicity (3%), 1 pt (0.5%) genitourinary toxicity. Two thirds of pts experienced toxicity before the adaptive phase and the third cycle of oxaliplatin, when prescribed.

Feasibility. One hundred fourty-nine pts completed RT with full Radiotherapy dose. Three cycles of Oxaliplatin were prescribed in 90 pts (59%), < 2 cycles in 62 pts (41%). Eighty-five percent of pts received > 80% of capecitabine dose.

Efficacy. All 152 pts underwent surgery, 5 pts (3%) abdominal-perineal resection, 147 (97%) pts conservative resection. Overall, 39 pts (26%) achieved pathological complete response, 26 pts (17%) TGR 1-5%, 22 pts (14%) TRG 6-10%. Considering pts who received 3 cycles of oxaliplatin, 26 out of 90 (29%) achieved pCR against 13/62 (21%) pts who received 1-2 cycles (not statistically significant). Ninety pts (60%) received adjuvant chemotherapy. Outcome. With a median FU of 50,3 months (5,9-149), relapse occurred in 45 pts (30%), locoregional in 5 pts (3%), distant in 43 pts (28%). Thirty-two pts (21%) died. Median DFS and OS were 105 and 133 months, respectively.

In the present protocol Authors aim at a further improvement of pCR rate by increasing the RT dose to residual tumor in patients selected by ERI_TCP and by better identification of the residual tumor.

Intervention description

Radiotherapy

Initial RT phase (standard phase) Patients are immobilized on Comby-Fix® in supine position. Simulation CT and T2 weighted RM of the pelvis without contrast media are performed and matched.

Clinical Target Volume (CTV) definition follows international guidelines and encompass primary tumor, mesorectum, obturator lymph-nodes, internal iliac and common iliac chains as well as the whole anterior surface of sacrum, coccyx and piriformis muscle. Anal canal and perineum are included in case of lesser limit of tumor located less than 6 cm from the anal verge. External iliac lymph-nodes are included in case of T4 disease infiltrating pelvic organs or wall. Inguinal lymph-nodes are treated in case of massive infiltration of the anal canal.

Planning target volume (PTV) is defined as CTV expanded by 0.5 cm in all directions. The prescribed dose to PTV is 27.6 Gy in the first 12 out of 18 fractions (2.3 Gy/fr).

Adaptive RT phase (experimental phase)

After 8-10 fractions of RT a new CT (CTadapt) and T2 weighted RM (MRadapt) of the pelvis without contrast media are performed following the same set-up as the original planning scans and matched.

Adaptive Gross Tumor Volume (GTV) 1 (GTV1adapt) is defined as the portion of rectum included between the cranium-caudal limits of residual tumor. GTV2adapt is defined as the true residual tumor (T) still detectable on the high quality MR images with the help of an expert radiologist. PTV1adapt and PTV2adapt are obtained by expanding GTV1adapt and GTV2adapt of 0.5 cm in all directions.

ERI_TCP calculation

In order to confirm the enrollment of the patient into the study, ERITCP at half-RT needs to be assessed. For this reason, tumor volumes are contoured on MRI T2 images by the same radiation oncologist (or radiologist) expert in rectal tumor MR imaging on axial images of initial MR (Vpre corresponding to GTVpre) and MRadapt (Vmid corresponding to GTV2adapt), and ERITCP is calculated as follows:

ERITCP=-ln[(1-(Vmid/Vpre))Vpre

The "hyperboost" to PTV2adapt will be delivered only in two cases:

1. the inferior margin of PTV2adapt, examined by multidisciplinary team composed of radiologists, surgeons and radiation oncologists, is located more than 10 mm from the hypothetical resection line
2. ERITCP < 32.6, defining responsive patients

Dose prescription for adaptive phase PTV: 2.3 Gy/F x 6 fractions (total dose 41.4 Gy in 18F) PTV1adapt : 3.1 Gy/F x 6 fractions (total dose 46.2 Gy in 18 fractions). PTV2adapt : 4.0 Gy/F x 6 fractions (total dose 51.6 Gy in 18 fractions)

Treatment Planning and image-guidance; preliminary planning study Patients are treated using Volumetric Modulated Arc Therapy or Tomotherapy. Concerning planning optimisation, PTVS coverage and dose homogeneity have the highest priority: the fraction of PTV/PTV1-2adapt receiving 95% of the prescribed dose (V95%) is set ≥98% and maximum PTV/PTVadapt dose (Dmax) <105% respectively, both in the initial and in the adaptive plan; the homogeneity of the dose distribution within PTV is minimized even in the adaptive plan. Dmax of femoral heads is set <42Gy. Bladder and bowel are spared "as much as possible" in the portion outside PTV. The dose to external genitalia are minimized by blocking any beam passage through them.

In order to better define proper templates for plan optimization with the hyperboost (delivering up to 4Gy/F to the true residual tumor), an in-silico plan comparison study on at least 10 previously treated patients showing ERITCP<32.6 will be accomplished. Plans with hyperboost and without (i.e.: the clinically delivered plans) will be compared and evaluated in terms of potential worsening of dose-volume parameters of the different organks at risk, primarily bowel and bladder.

Concerning image-guidance, before delivering each fraction, a ConeBeam CT (CBCT) or Megavoltage CT (MVCT) scan is performed: first, an automatic bone match between planning kVCT and CBCT or MVCT is carried out; the physician then apply fine manual adjustments, if necessary, in order to minimize the residual error after bone matching due to rotations and/or deformations occurring during therapy.

Concomitant chemotherapy

Concomitant chemotherapy consists of Oxaliplatin 85 mg/m2 delivered on days -14, 0, +14, and Capecitabine 825 mg/m2 twice daily or less according to DPYD status, assumed from day -14 until the end of RT. Radiotherapy starts on day 0.

Dose reduction In case of toxic effects, chemotherapy cycles will be delayed until the neutrophil count will be 1·5 × 109 cells per L or higher and the platelet count was 100 × 109 per L or higher. Briefly, dose reductions will be recommended in patients who develop febrile neutropenia, grade 4 neutropenia lasting more than a week, grade 3-4 thrombocytopenia, or any toxicity worse than grade 2. The oxaliplatin dose will be reduced when grade 2 peripheral neuropathy will be painful or persistent between cycles and oxaliplatin will be stopped in case of motor impairment. Capecitabine dose will be adapted to blood counts and renal clearance. Dose reduction or discontinuation of capecitabine will be considered in case of grade 2 toxicity.

Statistics

Description of Statistical Methods Sample size is calculate with the Fisher test for a single proportion. The primary endpoint will be the proportion of patients achieving pCR among those with ERITCP <32.6. If the 70% target is not reached, statistical significance will be assessed using an exact binomial test comparing the observed proportion to the expected proportion, with calculation of the p-value and 95% confidence interval.

Secondary endpoints will be described as frequencies and percentages (acute toxicity, anal sphincter preservation, surgical complications). Survival outcomes (DFS, LRFS, DRFS, OS) will be analyzed using the Kaplan-Meier method, compared with the log-rank test, and, when appropriate, evaluated in multivariable analysis using Cox proportional hazards regression to estimate hazard ratios with 95% confidence intervals.

Sample Size Determination This is a prospective single-center study. The historical pCR rate is 41% in patients with ERI_TCP < 32.6 as reported from our previous data [16]. Infact in that study, the overall pCR rate reported among 63 patients was 31.7%. When excluding the quartile of patients with ERI_TCP >32.6-who are known to have an extremely low likelihood of achieving pCR-the rate increased to 41%, which we therefore adopted as the reference value for this subgroup. The primary endpoint is to demonstrate of an increase in the pCR from the current 41% to 70% in the patient with ERITCP <32.6. Using Fisher's exact test for a single proportion, a sample size of 22 patients was estimated to provide 80% power to detect a difference from the historical value (α = 0.05, β = 0.20). Considering that approximately 25% of enrolled patients are expected to have an ERI_TCP > 32.6, it was estimated that enrolling 30 patients in total will be sufficient to demonstrate the primary endpoint. Accounting for a potential <10% rate of patient refusal and/or inability to undergo surgery, a total sample size of 33 patients is considered adequate to complete the study (20).

Patients who refuse or are not fit for surgery will be considered not evaluable for the primary end point but evaluable for secondary endpoints