Contact X-ray Brachytherapy as a sole treatment in selected patients with early rectal cancer – Multi-centre study

Highlights • We have evaluated the role of sole Contact X-ray brachytherapy (CXB) as a treatment option for early rectal cancer.• There have been no recently published analyses of the efficacy of CXB as a sole modality treatment for early rectal cancer.• A high clinical complete response rate (82%), low local regrowth rate (18%), and few distant tumour relapses were observed.• We observed poorer outcomes in patients who had received prior pelvic radiotherapy for other cancers.

Over the last decade, increased detection of early-stage rectal cancers through national bowel cancer screening programmes [7,8], a higher incidence of bowel cancer in the elderly [9], and the increased prevalence of medical comorbidities associated with advancing age [10], support the use of organ-preserving treatments.These include transanal excision (TAE) or endoscopic mucosal resection (EMR) with or without adjuvant radiotherapy, and non-surgical treatment modalities including external beam (chemo)radiation (EBRT/EBCRT), or exclusive endocavitary radiotherapy [11][12][13][14][15][16].Adopting these more conservative strategies aims to minimise post-surgical complications, decrease mortality rates, and enhance patients' overall quality of life [17,18].
Since its introduction as a rectal cancer treatment, Contact X-ray Brachytherapy (CXB) has been used as a standalone therapy for very early-stage tumours and can be supplemented by the incorporation of EBRT or interstitial brachytherapy using Iridium ( 192 Ir) [19,20].Sole CXB treatment remains an attractive option for early-stage rectal cancers in view of the low likelihood of lymphatic metastasis in T1/T2 tumours (12-22 %) [21,22].
Appelt et al. [23] demonstrated a clear dose-response relationship in patients with locally advanced rectal cancer who underwent neoadjuvant chemoradiation.The therapeutic efficacy of radiation relies on balancing tumour control and minimising side effects in normal tissues.Despite advanced precision in external radiotherapy techniques, achieving accurate tumour irradiation without damaging nearby healthy tissues remains challenging [24].CXB has the advantage of delivering a high radiation dose directly to the tumour (20-30 Gy/session), with a rapid dose fall-off as tissue depth increases due to its low energy (50kVp), (50 % of the dose at 5 mm depth and 30 % at 10 mm depth), thereby minimising damage to the surrounding normal tissues [25,26].
Although the oncological outcomes following CXB alone may not be comparable to the standard surgical results [27], it offers significant advantages for certain patients, particularly those who have limited capacity to undergo surgical procedures.CXB is an ambulatory treatment that generally does not require general anaesthesia, making it suitable for patients who are at high surgical risk due to advanced age or multiple comorbidities.Additionally, patients can resume their daily activities soon after treatment [26,27].These result in lower overall treatment costs for this therapy compared to major surgical procedures [28].Furthermore, CXB treatment due to its limited penetration properties, avoids disruption of the surgical meso-rectal plane.This is an important advantage over local excision, which can impact post-surgical morbidities if subsequent salvage completion surgery is required following local treatment failure [29,30].
Although numerous studies over the past seven decades have examined the outcomes of CXB alone, there has been a lack of recent multi-centre reports with meaningful numbers of patients.We therefore conducted this multi-institutional study to evaluate the efficacy of sole CXB treatment with curative intent, particularly in the older population who were considered high surgical risk or those who declined surgery for various reasons.We also assessed the prognostic factors influencing oncological outcomes and the tolerability of treatment.

Patient selection
This retrospective multi-centre study utilised data from the Guildford Colorectal Database System, a prospectively maintained national database employed by four UK CXB treatment centres: St. Luke's Cancer Centre (Guildford), Clatterbridge Cancer Centre (Liverpool), Nottingham City Hospital, and Queen's Oncology Centre (Hull), as well as Uppsala Cancer Unit in Sweden.The study included all consecutive patients with well/moderately differentiated, (cT1-2,N0,M0) rectal adenocarcinoma who received CXB treatment alone with curative intent between 2009 and 2021.Patients with more advanced tumour stages (cT3/T4,N1, or M1) were excluded, as CXB alone is unlikely to cure such cases due to its limited penetration into deeper tissues.We also excluded patients who received CXB alone as adjuvant treatment after local excision or for luminal regrowth during watch-and-wait after neoadjuvant (chemo)radiation, as these cases have been analysed separately [12,31].After obtaining approval from all participating institutions, we collected data on patient and tumour characteristics/ staging, treatment details, and follow-up information.For missing data not recorded in the system, we retrieved information where possible from local patient records.

Patient assessment and CXB treatment
Baseline tumour characteristics and staging were evaluated through clinical examination, endoscopic assessment, histological confirmation, cross-sectional imaging (magnetic resonance imaging (MRI)), whichwas mainly used for determining tumour and nodal stage unless it was contraindicated (e.g.patient with a cardiac pacemaker) and computed tomography (CT) (chest and abdomen) to exclude any distant metastases.FDG-PET/CT was not routinely performed unless it was needed to evaluate an equivocal finding on a contrast-enhanced CT or MRI scan or  Before starting CXB treatment, patients were reviewed at the local CXB multi-disciplinary team (MDT) meeting to determine the suitability and intent of treatment.Each patient was counselled to fully understand that CXB alone is not a standard of care for early-stage rectal cancer and that they may need to consider surgery if there is residual tumour or local regrowth after CXB.All patients provided informed consent before treatment commencement.
CXB was delivered as an outpatient treatment using the Papillon-50 ©machine (50kVp X-rays (HVL 0.64 Al, 2.7 mA), Ariane, Alfreton, UK), 20-30 Gy per fraction delivered every 2 weeks apart, through a rectal treatment applicator (size 30, 25, or 22 mm) at a focal source surface distance of 29, 32, or 38 mm respectively.A standard dose of 90 Gy (rectal mucosal surface dose) was delivered in 3 fractions over 4 weeks, with a fourth 20 Gy dose (total 110 Gy) being administered to selected patients who had minimal visible and/ palpable tumours still present after the third fraction.

Follow-up
Patients who achieved a complete/near Clinical Response (cCR/ nCR) at 8-12 weeks following completion of treatment underwent regular assessments, including digital rectal examination, flexible sigmoidoscopy or rectoscopy at their local centre and CXB centre alternatively, and MRI scans at subsequent 12-14 week intervals during the first 2 years where the likelihood of recurrence is highest.Subsequently, evaluations were conducted every 6 months in the 3rd year.Usually, only endoscopic examinations were performed in the 4th and 5th years.Patients who did not achieve cCR/nCR, which was confirmed by triple assessment (DRE, endoscopy, and MRI) with/without histological confirmation, by 24 weeks were considered for surgical treatment options if they were agreeable and fit enough.

Outcome measures
Oncological outcomes were evaluated for the whole group and subgroups based on tumour stage (T1 versus T2) and the reasons for undergoing sole CXB without involving EBRT.The patients were divided into 3 groups as follows; those patients who were fit enough for surgery but refused (group A), those who had a high risk for surgery due to age and/or medical comorbidities (group B), and those who had a history of prior pelvic radiotherapy for other malignancies including prostate (n = 13), gynaecological (n = 6), and bladder (n = 1) cancers (group C).The primary endpoints included cCR, local control rate (LCR), disease-free survival (DFS), and overall survival (OS).Disease-free survival for patients who have achieved cCR was calculated from the date of the last radiotherapy treatment to the date of locoregional recurrence after R0/ R1 resection of the primary tumour, non-salvageable local regrowth/R2 salvage resection, the occurrence of a second primary, distant metastasis, or last follow-up.For those who had undergone salvage surgery for their residual disease, disease-free survival was from the date of R0/R1 resection of the primary tumour to the date of locoregional recurrence, the occurrence of a second primary, distant metastasis, or last follow-up.Overall survival was defined as the period from the date of diagnosis to the date of the last data review or death from any cause.Secondary outcomes involved assessing the influence of patient and tumour factors on oncological outcomes and post-CXB radiation toxicities using Common Terminology Criteria for Adverse Events (CTCAE) criteria version 5.0 [36].

Statistical analysis
Quantitative data were expressed as medians with interquartile ranges (IQR) or means with standard deviations (SD), while categorical data were reported as counts with percentages.Categorical variables were compared using the χ2 test, with Fisher's exact test (two-tailed) employed when necessary.Continuous variables, which were not normally distributed, were compared using the independent-samples Mann- Whitney U test.Survival differences were examined with Kaplan-Meier curves, and the Log-rank test was used for statistical assessment.Associations between tumour characteristics and survival risks were analysed using Cox proportional hazards models, while logistic regression was used to evaluate binary margin outcomes.A p-value of < 0.05 was considered statistically significant.Statistical analyses were conducted using R version 4.3.2.

Results
From 2009 to 2021, 140 patients in the database were treated solely with CXB.However, only 76 patients were eligible for the study after excluding those who had more advanced stages (cT3-4/N1/M1) (n = 39), individuals not having accurate staging (cTx/Nx/Mx) (n = 8), patients who had been lost to follow-up (n = 14) and those who underwent combined treatment with high-dose-rate brachytherapy (n = 3).Oncological and clinical outcomes were analysed for the entire group and sub-groups based on the tumour stage and the reasons for administering CXB alone.Detailed study characteristics are presented in Fig. 1.

Whole group analysis
With a median follow-up of 26(IQR: 12-49) months, the median age of patients was 78(IQR: 67-84) years and the majority (82 %) were aged between 65-98 years.Approximately 47 % of patients, despite being apparently fit, declined surgery.The mean tumour size was 2.4 ± 1.0 cm.Demographic data for the whole group and subgroups are shown in Table 1.

Predictive factors
Patients with a history of previous pelvic radiotherapy were found to have a lower likelihood of achieving cCR.The patient's age, performance status, high risk for surgery, and history of prior pelvic radiation were significant predictors for overall survival (Supplementary Table S1).

Follow-up and salvage treatments
Among the 14 patients who had residual disease following CXB, 5 opted for salvage surgeries: abdominoperineal resections (APER) = 3, low anterior resection (LAR) = 1, and transanal endoscopic microsurgery (TEMS) = 1, with all achieving R0 resections except R1 resection in the TEMS case.Two patients declined surgery and instead received the best supportive care, while the remaining 7 were considered high surgical risk; of these, 2 underwent external beam radiotherapy (EBRT), 1 received palliative chemotherapy, and 4 were given the best supportive care.
Of the 11 patients who experienced local tumour regrowth, 4 underwent APER (all R0 resections), and 1 patient, although fit enough, refused further treatment.The remaining 6 patients were unfit for surgery: received palliative radiotherapy (EBRT and further CXB) and 4 received the best supportive care.Among the two patients who experienced pelvic relapse, one underwent salvage surgery, while the other, who was unfit for surgery, received the best supportive care.

Post-radiation toxicities
Toxicity profiles were available for 47 out of the 76 patients.The predominant post-treatment symptom was late G1-2 rectal bleeding; observed in 12 patients (26 %).Grade 3 bleeding, necessitating argon beam treatment, occurred in only two patients (4.2 %).Erratic bowel, rectal pain, and urgency suggestive of acute proctitis symptoms (G1-2) manifested in 4 patients (9 %).None of the patients reported symptoms suggestive of impaired anal sphincter function.

Discussion
In this study, the overall cCR rate was 82 % (range: 70-93) while previous studies that evaluated the efficacy of CXB alone treatment between 1951 and 2006 [13,14,[32][33][34][35][36][37][38][39][40][41][42], reported a cure rate of 63-95 %  (with 5-20 % rate of residual disease).The local regrowth rate in our study was also comparable to those reported in previous CXB alone studies (18(8-29)% vs 5-28 %).The distant relapse rate was much lower in our study than previous findings (3 % vs 6-13 %), and no patients experienced nodal relapse, whereas one previous study reported a 7 % nodal relapse rate [13].The 5-year DFS rate was 66(48-78)% vs 66-97 % in the historic series, and the OS rate was 58(44-75)% vs 48-84 %. (Supplementary Table S2) The findings from our study are therefore generally consistent with the results published in these historic series.However, earlier studies utilised various staging systems and methods for assessing response, primarily using DRE and TRUS/ERUS.The development of modern imaging technologies for more accurate staging since those studies were conducted, could potentially introduce bias when directly comparing historical results with those found in our study.
Habr-Gama et al demonstrated cCR rates ranging from 57-86 %, locoregional recurrence rates between 20-31 %, and distant relapse rates of 11-26 % in a cohort of 81 patients with early rectal cancers (cT2N0M0) who were treated with standard (50.4Gy) and extended chemoradiation (54 Gy) [15].The risk of local recurrence after local excision alone, without adjuvant radiotherapy, for T1 tumours was 4-13 % and 19-29 % for T2 tumours [43,44].The heterogeneous nature of patient characteristics in our cohort limits direct comparisons of DFS and OS between our study and these other organ-preserving studies.Despite these constraints, the local control rate reported in our cohort was comparable to those observed with these organ-preserving options, while achieving a lower incidence of distant relapse.Patients who had T2 tumours, a history of previous pelvic radiotherapy, and who were at high risk for surgery, exhibited lower rates of cCR.However, no statistical significance was observed, which may be attributed to the relatively small cohort sizes.In contrast, patients who were high-risk-for-surgery showed a significantly lower risk of local regrowth than the other groups, thereby achieving a higher rate of DFS.Those patients who had previous radiation exposure for their previous unrelated type of tumour also experienced an increased risk of local tumour relapse, and lower DFS and OS.This may be related to the presence of radioresistant tumour clones and a greater proportion of T2 tumours within this group.Overall, we found no significant differences in outcomes between patients with T1 and T2 tumours, similar to observations in previous studies.However, the results of our statistical analysis may have been influenced by the small sample size of our cohort.
Previous radiation exposure was identified as an adverse predictor for local tumour control and OS, which could be attributed to the potential development of radioresistant tumour clones [45].Moreover, in this group, CXB alone was used in slightly higher-risk settings (such as T2 tumours) where a combination of EBRT and CXB would probably have been recommended if previous radiation had not been administered to this area.
No serious adverse effects were reported in this study, with comparable rates of self-limiting radiation toxicities, similar to other studies [35,46].No patients reported symptoms indicative of impaired anal sphincter function, a common post-treatment side effect of TAE and (chemo)radiation [17,47].
This study has several limitations.The data primarily relied on a database that was designed to register details such as primary treatment, toxicities, and surgical information.Unfortunately, some information, particularly long-term outcomes for patients who underwent salvage therapies following CXB failures at certain centres, was not recorded.These patients were returned to the care of colorectal surgeons who did not participate in the project, limiting our ability to thoroughly assess the feasibility of salvage treatment and the long-term outcomes for these patients.Moreover, the relatively small cohort size and uneven distribution of patients between sub-groups might have impacted the statistical comparison of outcomes.Additionally, the considerable heterogeneity among patient factors within our cohort may have introduced potential bias when assessing the results of sole CXB and comparing these with outcomes from other treatment options.Therefore, we recommend conducting prospective studies that adequately control for patient and tumour confounding factors to provide a more robust understanding of the role of sole CXB treatment compared to various other organ-preserving treatment options.

Conclusions
Our study confirmed a high rate of cCR, satisfactory local tumour control, and disease-free survival in a modern cohort of early rectal cancer patients (cT1/cT2, cN0, cM0) who received CXB alone.The lower overall survival rate likely reflects the characteristics of the elderly and comorbid cohort in our study population.Inferior oncological outcomes were also observed in patients with a history of prior pelvic radiotherapy.Nevertheless, considering its several advantages over other organ-preserving treatment options, CXB alone can be considered a viable alternative for selected patients.

Fig. 2A .
Fig. 2A.Flow diagram illustrating the oncological outcomes of sub-groups based on T-stage.

Fig. 2B .
Fig. 2B.Flow diagram illustrating the oncological outcomes of sub-groups based on reasons for having CXB alone.Group A: those fit for surgery but refusing surgery, Group B: those at high risk for surgery due to advancing age/medical comorbidities, Group C: those with a history of previous pelvic radiotherapy, cCR: clinical Complete Response, APER; Abdomino Peritoneal Excision of Rectum, LAR: Low Anterior Resection, TEMS: Transanal Endoscopic Mucosal Excision, BSC: Best Supportive Care, NA: not available information for salvage treatment.

Fig. 3A .
Fig. 3A.Kaplan-Meier curve for disease-free survival of the whole group.Fig. 3B.Kaplan-Meier curve for disease-free survival based on T-stage.

Fig. 3C .
Fig. 3C.Kaplan-Meier curve for disease-free survival based on reasons for having CXB alone.

Fig. 4A .
Fig. 4A.Kaplan-Meier curve for overall survival of the whole group.

Fig. 4B .
Fig. 4B.Kaplan-Meier curve for overall survival based on T-stage.Fig. 4C.Kaplan-Meier curve for overall survival based on reasons for having CXB alone.

Table 1
Patient and tumour characteristics of the whole and sub-groups.

Table 2
Oncological outcomes of the whole and sub-groups.