Colorectal cancer survival in patients without curative measures: a retrospective cohort study
Aman Saini#
, Ademola Adeyeye#, Hao Gao, Molly Steele, Alisha Jafri, Pranay Ruparelia, Bhavya Manchukonda, Amyn Haji
Contributions: (I) Conception and design: A Saini, A Haji, A Adeyeye; (II) Administrative support: A Saini; (III) Provision of study materials or patients: A Saini, A Haji, A Adeyeye, H Gao; (IV) Collection and assembly of data: A Saini, H Gao, M Steele, A Jafri, P Ruparelia; (V) Data analysis and interpretation: A Saini; (VI) Manuscript writing: All authors; (VII) Final approval of manuscript: All authors.
#These authors contributed equally to this work as co-first authors.
Background: A substantial proportion of colorectal cancer (CRC) patients are managed without curative intent due to frailty, comorbidity, or patient choice. Outcomes in this heterogeneous population remain incompletely characterised. We evaluated survival and treatment patterns in patients managed non-curatively, including exploratory analyses by tumour sidedness and metastatic status.
Methods: We conducted a retrospective cohort study of patients diagnosed with CRC at King’s College Hospital, London, between 2013 and 2022. Patients with histologically or radiologically confirmed CRC discussed at the multidisciplinary team (MDT) meeting were included; those receiving curative-intent treatment were excluded. Baseline variables included Eastern Cooperative Oncology Group (ECOG) performance status, tumour location, and metastatic disease at diagnosis. Overall survival (OS) from diagnosis was analysed using Kaplan-Meier methods and multivariable Cox regression.
Results: Of 1,501 patients with CRC, 267 (18%) were managed non-curatively. Mean age was 74.7 years, and 60.3% were male; 59.9% were ECOG 0–2. Chemotherapy (28%) and surgery (15.7%) were the most frequent treatments. 5.2% required emergency surgery. Median OS was 210 days [95% confidence interval (CI): 205–210]. On multivariable analysis, ECOG ≥3 [hazard ratio (HR) 2.40, 95% CI: 1.42–4.05; P=0.001] and metastatic disease (HR 2.36, 95% CI: 1.41–3.94; P=0.001) were independently associated with worse OS. Tumour sidedness was not associated with survival (HR 1.00, 95% CI: 0.62–1.61; P=0.99).
Conclusions: Patients with CRC managed non-curatively are often ambulant and may continue to receive supportive interventions aimed at maintaining quality of life. These findings support individualised risk stratification and prospective evaluation incorporating frailty and quality-of-life measures.
Keywords: Colorectal cancer (CRC); palliation; frailty
Submitted Aug 18, 2025. Accepted for publication Mar 26, 2026. Published online May 18, 2026.
doi: 10.21037/jgo-2025-670
Highlight box
Key findings
• In this single-centre retrospective cohort of 267 patients with colorectal cancer (CRC) managed with non-curative intent, many patients retained functional capacity, with 59.9% having Eastern Cooperative Oncology Group (ECOG) performance status 0–2 at diagnosis. Median overall survival was 210 days. On multivariable analysis, metastatic disease and impaired functional status (ECOG ≥3) were independently associated with worse survival, while tumour sidedness was not.
What is known and what is new?
• A substantial proportion of CRC patients are managed without curative intent, but this population is heterogeneous and incompletely characterised in large registry datasets, which often lack granularity on functional status and decision-making.
• This study provides detailed real-world data from a multidisciplinary team setting, demonstrating that patients classified as non-curative are not uniformly end-stage. Many remain ambulant and receive active interventions. Importantly, survival appears driven by metastatic burden and functional status rather than tumour location, suggesting that traditional prognostic frameworks may not fully apply in this population.
What is the implication, and what should change now?
• Non-curative intent should not be equated with terminal care at diagnosis. Clinical decision-making should incorporate structured assessment of functional status and frailty, rather than relying solely on disease factors.
• Future research should prioritise prospective studies integrating frailty metrics and quality-of-life outcomes to better stratify risk and guide personalised treatment pathways in this growing patient population.
Introduction
Colorectal cancer (CRC) is a leading cause of cancer-related morbidity and mortality worldwide. Surgical resection remains the cornerstone of curative management (1); however, an increasing proportion of patients are managed non-curatively due to advanced disease, comorbid conditions, or personal preference. With an ageing population, particularly in developed countries, the number of patients considered unsuitable for curative treatment is rising (2).
Large national databases such as Surveillance, Epidemiology, and End Results (SEER) and the National Cancer Database have provided extensive epidemiological insights into prognostication and treatment outcomes for such patients with advanced CRC, but lack detail concerning functional status, frailty, multidisciplinary team (MDT) decision-making, and the rationale for assigning non-curative intent. These factors are central to real-world treatment allocation and prognostic assessment yet are rarely captured in national cancer registries. Patients deemed non-curative by MDTs also represent a heterogeneous group; many do not have metastatic disease but are considered too frail, too comorbid, or unable or unwilling to undergo major surgery. This clinically important subset remains poorly described in the literature (3).
Single-centre cohort studies can complement national datasets by providing granular, real-world information about functional capacity, hospital utilisation, and the patterns of care that shape MDT decision pathways. Understanding these characteristics is increasingly relevant as healthcare systems seek to anticipate the needs of an ageing cancer population and to optimise resource allocation, supportive care, and palliative interventions.
This study aims to characterise the clinical profile, functional status, treatment patterns, and outcomes of patients with CRC who were managed non-curatively at a tertiary cancer centre with MDT review. Using Eastern Cooperative Oncology Group (ECOG) performance status as a quantitative measure of baseline function, we hypothesised that a proportion of patients labelled non-curative retain measurable functional capacity and continue to receive active interventions. We aim to inform clinical decision making and highlight areas where prognostic tools and supportive care pathways may require refinement. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-670/rc) (4).
Methods
Study design and setting
This is a retrospective cohort study conducted at King’s College Hospital (KCH), London, a tertiary referral centre for CRC. Data were collected from 2013 to 2022.
Inclusion criteria
We included all adult patients with a histologically or radiologically confirmed diagnosis of CRC who were documented as being managed with non-curative intent. Non-curative intent was defined as a documented MDT decision that curative resection was not appropriate, due to advanced disease, comorbidities or frailty precluding major surgery, or patient preference. This included patients receiving palliative systemic therapy or surgery, as well as those managed with best supportive care alone.
Data sources
Data were obtained from a prospectively maintained database curated by the KCH specialist colorectal team, contributing to the National Cancer Audit. Supplementary data were retrieved from electronic medical records. Clinical terminology and coding followed the NHS Data Dictionary standards (5).
Outcomes
The primary outcome was overall survival (OS) from the time of diagnosis. The secondary outcomes included: type and frequency of treatments received, Hospital admission rates and duration of stay as well as performance status at diagnosis.
Statistical analysis
Descriptive statistics were used to summarise patient demographics, tumour characteristics, and treatment modalities. OS was defined as the time from date of diagnosis to date of death from any cause. Patients alive at the end of follow-up were right-censored at 23 January 2025.
Patients with no recorded follow-up beyond the date of diagnosis were excluded from survival analysis, as they contributed no time-at-risk and would otherwise be censored at time zero, a scenario known to distort Kaplan-Meier estimates.
Baseline characteristics at diagnosis were compared between patients included in survival analysis and those excluded due to absence of information following diagnosis, to assess potential selection bias. Continuous variables were compared using two-sample t-tests, and categorical variables using Chi-squared or Fisher’s exact tests, as appropriate. Survival distributions were estimated using the Kaplan-Meier method and compared using the log-rank test. Median survival with 95% confidence intervals (CIs) was calculated.
Univariable and multivariable Cox proportional hazards regression models were used to estimate hazard ratios (HRs) and 95% CIs for factors associated with OS. Covariates were selected a priori based on clinical relevance and potential confounding and included age, sex, ECOG performance status (0–2 vs. ≥3), metastatic status (M0 vs. M1), and primary tumour location (left-sided vs. non-left). Analyses of tumour sidedness and metastatic status in relation to OS were restricted to patients included in the survival analysis cohort (n=89). The proportional hazards assumption was assessed using Schoenfeld residuals and was not violated.
All statistical tests were two-sided, and a P value <0.05 was considered statistically significant. Analyses were performed using Stata 13.0.
Ethical approval
This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This retrospective cohort study utilised anonymised data collected as part of a registered service evaluation at King’s College Hospital NHS Foundation Trust. Under UK Health Research Authority guidance, formal Research Ethics Committee approval and reference number were not required for this type of study. Institutional approval was obtained from the Director of Surgery at King’s College Hospital NHS Foundation Trust. As no identifiable patient data were used and there was no direct patient contact, individual patient consent was not required.
Results
Cohort characteristics
Out of 1,501 CRC patients, 267 (18%) met inclusion criteria. The mean age was 74.7 [standard deviation (SD) 13.8] years; 60.3% were male. At diagnosis, 59.9% of patients were ambulant for at least 50% of the day. Performance status, tumour location and metastatic status details are shown in Tables 1-3. Patients included in survival analysis (n=89) were similar at baseline to those excluded due to lack of follow-up (n=178) with respect to age, sex, performance status, metastatic status at diagnosis, and tumour site (P>0.05). Baseline characteristics are summarised in Table S1. To allow readers to assess baseline differences between patients managed with non-curative intent and the wider CRC population, we have added a supplementary table presenting baseline clinicopathological characteristics stratified by study inclusion status in Table S2.
Table 1
| Variable | Value |
|---|---|
| Age at diagnosis, years | 74.7±13.8 |
| Sex | |
| Male | 161 (60.3) |
| Female | 106 (39.7) |
| Performance status (n=260) | |
| Able to carry out all normal activity without restriction | 15 (5.9) |
| Restricted in strenuous activity but ambulatory and able to carry out light work | 60 (23.1) |
| Ambulatory and capable of all self-care but unable to carry out any work activities; up and about more than 50% of waking hours | 85 (32.7) |
| Symptomatic and in a chair or in bed for greater than 50% of the day but not bedridden | 76 (29.2) |
| Completely disabled; cannot carry out any self-care; totally confined to bed or chair | 24 (9.2) |
| Site of cancer, ICD-10 major classification (n=267) | |
| Sigmoid colon (flexure) | 68 (26.7) |
| Rectum | 57 (21.3) |
| Caecum (Ileocaecal valve) | 39 (14.7) |
| Ascending colon | 27 (10.1) |
| Hepatic flexure | 21 (7.9) |
| Transverse colon | 17 (6.4) |
| Descending colon | 12 (4.5) |
| Anus/anal canal | 8 (3.0) |
| Rectosigmoid junction | 8 (3.0) |
| Splenic flexure | 7 (2.6) |
| Appendix | 3 (1.1) |
| Staging of cancer | |
| Tumour | |
| Tx | 71 |
| T1 | 6 |
| T2 | 42 |
| T3 | 91 |
| T4 | 57 |
| Node | |
| Nx | 74 |
| N0 | 53 |
| N1 | 88 |
| N2 | 51 |
| N3 | 1 |
| Metastasis | |
| Mx | 11 |
| M0 | 105 |
| M1 | 151 |
| Care plan intention with reason (n=267) | |
| Treatment, non-curative intention | 182 (68.2) |
| Unfit, not specified | 83 |
| Unfit, advanced disease | 62 |
| Unfit, comorbidity | 31 |
| Patient declined | 6 |
| No treatment | 85 (31.8) |
| Unfit, comorbidity | 37 |
| Unfit, advanced disease | 30 |
| Patient declined | 18 |
Data are presented as mean ± SD, n (%), or number. ICD-10, International Classification of Diseases, 10th Revision; SD, standard deviation.
Table 2
| Variable | Value |
|---|---|
| Non-curative treatments received | |
| Chemotherapy | 74 (28.0) |
| Radiotherapy | 22 (8.2) |
| Surgery, any | 42 (15.7) |
| Surgery, emergency | 14 (5.2) |
| Number of hospital admissions post diagnosis (n=29) | 2.2±2.7 |
| Average time spent in hospital per patient, days | 13.5±18.6 |
| Average transfusions per patient, red cell units | 0.33 |
| Average intravenous iron transfusions per patient, episodes | 0.31 |
| ICU admissions | 0 |
| Median survival from diagnosis, days (n=89) | 210 [205–210] |
| Recorded places of death (n=5) | |
| Hospital | 1 |
| Hospice | 1 |
| Usual home residence | 3 |
Data are presented as n (%), mean ± SD, median [95% CI], or number, unless otherwise specified. CI, confidence interval; ICU, intensive care unit; SD, standard deviation.
Table 3
| Characteristic | Left-sided CRC (n=55) | Non-left-sided CRC (n=34) | P value |
|---|---|---|---|
| Age at diagnosis, years | 72.1±14.6 | 72.8±13.6 | 0.83 |
| Sex | 0.23 | ||
| Male | 33 (60.0) | 16 (47.0) | |
| Female | 22 (40.0) | 18 (53.0) | |
| ECOG performance status | 0.50 | ||
| 0–2 | 35 (63.6) | 24 (70.6) | |
| ≥3 | 20 (36.4) | 10 (29.4) | |
| Metastatic stage at diagnosis | 0.37 | ||
| M0 | 23 (41.8) | 11 (32.4) | |
| M1 | 32 (58.2) | 23 (67.6) |
Data are presented as mean ± SD or n (%). CRC, colorectal cancer; ECOG, Eastern Cooperative Oncology Group; M, metastasis; SD, standard deviation.
Disease staging and treatment intent
Of the 267 patients, 182 (68.2%) were allocated to active non-curative treatment plans due to comorbidities, advanced disease, or both. The remaining patients (31.8%) received no active treatment, mainly due to frailty or patient refusal.
Interventions and hospitalisation
Hospitalisation data were available for a subset of patients (n=29) with complete admission records. In this exploratory subgroup, there was a mean 2.2 hospital admissions per patient, with a mean 13.5-day length of stay. Chemotherapy (28%) and palliative surgery (15.7%) were the most frequent treatments (Table 2). Emergency surgery was required in 5.2% of cases. No patients required intensive care unit (ICU) admission. These findings are descriptive and should be interpreted cautiously given the limited sample size and potential selection bias.
Survival
Of the 267 patients, 178 had no recorded follow-up beyond the date of diagnosis and were excluded from survival analysis as these patients had no recorded survival time and could not contribute to time-to-event analysis. Among patients included in survival analysis (n=89), median OS from diagnosis was 210 days (95% CI: 205–210). Estimated OS was 53.9% (95% CI: 43.1–63.6%) at 6 months, 39.3% (95% CI: 29.2–49.3%) at 1 year, and 24.7% (95% CI: 16.3–34.0%) at 2 years. Maximum follow-up was 4,353 days. At the end of follow-up, 75 patients died and 14 patients were censored.
Kaplan-Meier analysis demonstrated significantly worse OS among patients with metastatic disease compared with those without metastases (log-rank P=0.01). In contrast, no significant difference in survival was observed according to primary tumour location (log-rank P=0.86). Kaplan-Meier survival curves are shown in Figures 1-3, including stratified analyses by tumour location and metastatic status.
On univariable Cox regression, metastatic disease was associated with increased mortality risk (HR 1.87, 95% CI: 1.14–3.07; P=0.01), as was ECOG performance status ≥3 (HR 2.11, 95% CI: 1.32–3.40; P=0.002). Age and sex were not significantly associated with survival in univariable analysis, and tumour location was not associated with mortality (HR 0.96, 95% CI: 0.60–1.53; P=0.86).
On multivariable Cox regression adjusting for age, sex, performance status, metastatic status, and tumour location, metastatic disease (adjusted HR 2.36, 95% CI: 1.41–3.94; P=0.001) and ECOG ≥3 (adjusted HR 2.40, 95% CI: 1.42–4.05; P=0.001) remained independently associated with worse OS. Age, sex, and tumour location were not independently associated with mortality. Results of univariable and multivariable analyses are presented in Table 5.
Table 5
| Variable | Univariable | Multivariable | |||
|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | ||
| Age (per year) | 1.01 (1.00–1.03) | 0.17 | 1.01 (0.99–1.03) | 0.37 | |
| Sex, male | 0.98 (0.62–1.54) | 0.91 | 1.02 (0.64–1.61) | 0.94 | |
| ECOG performance status ≥3 | 2.11 (1.32–3.40) | 0.002 | 2.40 (1.42–4.05) | 0.001 | |
| Left-sided CRC | 0.96 (0.60–1.53) | 0.86 | 1.00 (0.62–1.61) | 0.99 | |
| Metastatic disease | 1.87 (1.14–3.07) | 0.01 | 2.36 (1.41–3.94) | 0.001 | |
CI, confidence interval; CRC, colorectal cancer; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio.
Discussion
Summary of key findings
This study represents a contemporary single-centre cohort describing patients with CRC managed with non-curative intent following MDT review.
Despite their non-operative status, many patients retained reasonable performance status and received active treatment. The median survival was 210 days from diagnosis, which, although shorter than curatively treated patients, remains clinically meaningful. This is lower than the median result of 586 days of similar study of 33 patients with a similar level of frailty, with the caveat that their study excluded patients with metastatic disease (6); and the mean result of another study with similar exclusion criteria, which also excluded patients who passed away within 2 months of diagnosis (7).
Survival analysis demonstrated that metastatic disease and impaired functional status were the principal determinants of mortality in this cohort. On multivariable Cox regression, both metastatic disease and ECOG performance status ≥3 were independently associated with approximately a two-fold increase in the hazard of death, whereas age and sex were not independently predictive. These findings underscore the dominant prognostic importance of disease burden and functional reserve in patients managed with non-curative intent.
In contrast, primary tumour location was not associated with OS in either univariable or multivariable analysis. While biological and molecular distinctions between left- and right-sided CRCs are well established in curative and metastatic treatment settings, our findings suggest that, within a frail and predominantly palliative cohort, tumour sidedness does not independently influence survival once metastatic status and performance status are accounted for. This may reflect the overriding impact of systemic disease burden and physiological reserve in determining outcomes in this population.
Notably, the proportion of ambulant patients in our cohort challenges the assumption that non-curative intent equates to end-of-life care at diagnosis. The low ICU admission rate within our smaller cohort likely reflects appropriate case selection rather than denial of care, given the poor expected benefit of intensive care in frail cancer patients. However, hospitalisation data were available for only a small subset of patients and may not be representative of the wider cohort; these findings should therefore be interpreted as illustrative rather than definitive. These results demonstrate that patients considered unsuitable for curative surgery represent a heterogeneous group, some of whom continue to receive targeted oncological or supportive therapies. Future approaches should integrate frailty scoring and individualised treatment planning, perhaps leveraging endoluminal or minimally invasive techniques.
Limitations
This study is retrospective and may be affected by documentation gaps, particularly in more recent entries. The absence of structured quality of life measures limits conclusions on patient well-being. Exploratory hospital utilisation analyses were limited to a subset of patients with complete admission data, introducing potential selection bias and limiting generalisability.
The survival analysis was restricted to patients with recorded follow-up beyond the date of diagnosis and although baseline characteristics were comparable between included and excluded patients, this may introduce selection bias and therefore results should be interpreted as applying only to the smaller subgroup with available follow-up. The relatively modest sample size limits power to detect small differences in survival, particularly with respect to tumour location and subgroup analysis.
Future directions
Prospective studies incorporating frailty indices and validated quality of life tools are needed. Use of machine learning algorithms may help stratify risk and guide treatment allocation in this heterogeneous population. Further multicentre studies are warranted to clarify whether tumour sidedness has prognostic relevance in non-curative cohorts, or whether its apparent importance in other settings is attenuated in populations characterised by advanced disease and functional impairment.
Conclusions
This retrospective cohort study is a contemporary single-centre cohort describing patients with CRC managed with non-curative intent following MDT review and provides insight into patients’ clinical trajectories. Many retain functional independence and undergo active treatments, suggesting a role for nuanced, patient-centred care. Metastatic burden and functional status, rather than tumour location, were independently associated with survival in this population. Our findings underscore the need for improved prognostication tools and tailored therapeutic strategies in this expanding patient group.
Table 4
| Characteristic | M0 at diagnosis (n=34) | M1 at diagnosis (n=55) | P value |
|---|---|---|---|
| Age at diagnosis, years | 76.4±11.7 | 69.8±15.1 | 0.03 |
| Sex | 0.75 | ||
| Male | 18 (52.9) | 31 (56.4) | |
| Female | 16 (47.1) | 24 (43.6) | |
| ECOG performance status | 0.24 | ||
| 0–2 | 20 (58.8) | 39 (70.9) | |
| ≥3 | 14 (41.2) | 16 (29.1) | |
| Left-sided CRC | 0.37 | ||
| Yes | 23 (67.6) | 32 (58.2) | |
| No | 11 (32.4) | 23 (41.8) |
Data are presented as mean ± SD or n (%). CRC, colorectal cancer; ECOG, Eastern Cooperative Oncology Group; M, metastasis; SD, standard deviation.
Acknowledgments
None.
Footnote
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-670/rc
Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-670/dss
Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-670/prf
Funding: This research received no specific grant
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-670/coif). A.A. reports academic and research involvement in the development and evaluation of endoscopic electroporation technologies, including investigator-initiated studies related to Endoscopic Electroporation and related platforms. These technologies are being explored for the treatment of patients with colorectal neoplasms, as part of a PhD funded by a research grant from Mirai Medicals. A.H. reports academic supervision and clinical research involvement in studies evaluating novel endoscopic and oncological therapies conducted within his institutional role. The other authors have no conflicts of interest to declare.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This retrospective cohort study utilised anonymised routinely collected data as part of a service evaluation at King’s College Hospital NHS Foundation Trust. In accordance with UK Health Research Authority guidance, formal Research Ethics Committee approval and reference number were not required. Institutional approval was obtained from the Director of Surgery at King’s College Hospital NHS Foundation Trust. As no identifiable patient data were used and there was no direct patient contact, individual patient consent was not required.
Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
References
- NICE. Recommendations. Colorectal cancer. Guidance. NICE. 2021. Available online: https://www.nice.org.uk/guidance/ng151
- Cancer Research. Bowel cancer statistics. Cancer Research UK. 2025 [cited Mar 17, 2025]. Available online: https://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/bowel-cancer
- Sedrak MS, Freedman RA, Cohen HJ, et al. Older adult participation in cancer clinical trials: A systematic review of barriers and interventions. CA Cancer J Clin 2021;71:78-92. [Crossref] [PubMed]
- von Elm E, Altman DG, Egger M, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Int J Surg 2014;12:1495-9. [Crossref] [PubMed]
- NHS. NHS Data Model and Dictionary Service - NHS England Digital [Internet]. 2025 [cited Mar 17, 2025]. Available online: https://digital.nhs.uk/services/nhs-data-model-and-dictionary-service
- Franklyn J, Poole A, Lindsey I. Colon cancer survival in the elderly without curative surgery. Ann R Coll Surg Engl 2024;106:592-5. [Crossref] [PubMed]
- Bethune R, Sbaih M, Brosnan C, et al. What happens when we do not operate? Survival following conservative bowel cancer management. Ann R Coll Surg Engl 2016;98:409-12.
