Colorectal cancer is among the most common gastrointestinal malignancies. Current epidemiologic studies confirm the stabilization or the decrease of standardized incidence and mortality rates in high income countries, while a considerable increase in early-onset colorectal cancer is, also, noted (1,2). Furthermore, it is estimated that, over 300,000 new cases per year are diagnosed in Europe alone, while the pooled 5-year survival rate is approximately 65% (2-5). Additionally, almost 60% of these patients are older than 70 years old, with considerable comorbidities and physical impairments (3-5).

However, older patients (>65 years of age) are a very heterogeneous group, with notable deviations in terms of fitness and frailty (6-11). More specifically, the presence of underlying functional and mobility deficits combined with suboptimal nutritional status and ageism impact the prognosis of such patients (12-14). Therefore, a formal geriatric assessment with validated tools, generally provides more accurate classification, compared to a simple age cut-off point (12-15). These patients have been, traditionally, undertreated, receiving in many cases suboptimal oncological operations.

Additionally, due to the accompanying comorbidities and reduced survival expectancy, these patients are systematically excluded from clinical trials (6-11). Moreover, besides strict inclusion criteria, the lack of family support and the surgeons’ prejudice for increased risk of morbidity, led to a systematic under-representation of these patients in colorectal cancer studies (13,14). However, these disparities are quite problematic since older patients have the highest rates of malignancies, alongside major socioeconomic obstacles (13,14).

Surgical resection alongside perioperative therapy is the gold standard approach in colorectal cancer (7,9-11,16-18). However, radical resections are associated with a considerable morbidity and mortality burden. Thus, especially in older patients, frailty, and physical reserves limitations, often lead to the postponement or deviations from the optimal treatment (7,9-11,16-18).

Current evidence regarding the optimal management of older colorectal cancer patients, is still a matter of controversy (19-24). More specifically, initial cohorts suggested that older patients display a higher risk of perioperative complications, whereas recent studies reported that optimal oncological resections can be safely performed in this age subgroup (19-24). The significance of these is that optimal management of colorectal cancer in older patients has a direct impact on the oncological outcomes, including survival and recurrence.

Therefore, the aim of this prospective study was to compare the differences between older and younger colorectal cancer patients in terms of survival and the delivery of optimal surgical treatment.

Study endpoints. The primary endpoint of the study was comparison of the older (>70 year) and younger (<70 years) patient subgroups in terms of overall survival (OS) (25). Secondary endpoints included optimal surgery rates and OS comparisons based on optimal treatment and the interaction of optimal and age subgroups.

Optimal treatment was defined as the operation that allowed complete oncological clearance based on tumor location and stage. More specifically, for right and left colon cancer that was translated to colectomy adhering to the CME/CVL principles, with adequate resection margins. Similarly optimal treatment for rectal cancer was considered resection based on the TME principles followed by anastomosis when applicable. Any deviation from these principles, or any salvage operation (e.g., bypass procedures or defunctioning stomas) was considered as suboptimal treatment.

Study design. This prospective observational study was conducted at the Department of Surgery, University Hospital of Larissa, Larissa, Greece. All patients submitted to colorectal cancer surgery during the January of 2016 -April of 2020 period, were included in this study. Prior to patient inclusion, the appropriate local ethics committee approval and informed consent were received. The present study results are reported based on the STROBE guidelines (26).

Eligibility criteria. Eligible patients were considered all consecutive adult patients (³18 years), diagnosed with colorectal cancer and operated in our Department during the aforementioned period. The following exclusion criteria were considered: 1) non-adult patients, 2) refusal to participate and provide a signed informed consent, 3) operations performed outside the investigation period, 4) recurrent tumors and 5) patients not submitted or unfit for surgical treatment.

Treatment. All operations were performed by a group of 4 consultant surgeons. All surgeons had previous experience in open and laparoscopic colorectal surgery and had completed the respective learning curves. A standardized surgical technique was used in all cases. Dissection was performed with an energy source.

Data collection. Data regarding demographics [age, sex, American Society of Anesthesiologists -ASA score, Eastern Cooperative Oncology Group (ECOG) status, comorbidities, and Charlson Comorbidities Index (CCI)], tumor characteristics (TNM classification and tumor location) and operative and perioperative treatment outcomes (resection type, surgery delay, operation status, surgical approach, additional resections, and perioperative therapy) were recorded. Moreover, all perioperative adverse events, readmission and reoperation rates were recorded. Patient characteristics were retrieved from institutional databases, whereas morbidity and survival data were recorded during follow-up. The latter included scheduled outpatient visits and telephonic call evaluations.

Statistical analysis. Prior to statistical analyses, a Kolmogorov-Smirnoff normality test was performed. Chi-square test was applied in categorical variables, whereas a t-test and a Mann-Whitney U-test was used for normal and non-normal continuous outcomes, respectively. Pearson’s correlation coefficient was also introduced for association analyses. Survival analysis was based on the Kaplan-Meier curves and the Log-Rank test. Continuous data were provided in the form of Mean (Standard Deviation) or Median (Interquartile Range) on the basis of normality results, while categorical variables were reported as N (Percentage). Significance was considered at the level of P<0.05. Statistical analyses were completed by SPSS v.21 software.

Ethics approval. This study was performed in line with the principles of the Declaration of Helsinki.

Informed consent. Informed consent was obtained from all individual participants included in the study. Patients signed informed consent regarding publishing their data and photographs.

Patient characteristics. Overall, 166 patients were enrolled in the study (Table I). More specifically, 60 (36.1%) younger and 106 (63.9%) older patients were submitted to colorectal cancer surgery. Mean age in the former and the latter group was 62.27 and 77.8 years, respectively. There was no difference in terms of mean weight and sex allocation. Even though the older subgroup displayed a higher rate (p=0.007) of ASA II (56.7% vs. 61.3%) and ASA III (0% vs. 11.3%) patients, mean CCI scores were comparable (p=0.384) between the two groups. Older patients were not associated with increased rates of high ECOG (p=0.384) status. Similarly, besides coronary heart disease (1.7% vs. 12.3%), overall and specific comorbidities rates did not differ among the older and younger patients.

Treatment characteristics. In Table II, a summary of the TNM classification characteristics is displayed. In total, 35 cecum/ascending colon, 14 transverse colon, 60 descending colon and 57 rectal were introduced in this study. Moreover, 55 right colectomies, 2 extended right colectomies, 18 left colectomies, 24 sigmoidectomies, 40 low anterior resections, 3 abdominoperineal resections, 6 transverse colostomies, 8 loop ileostomies, 6 Hartmann’s procedures, 3 subtotal colectomies and 1 bypass were performed (Table III). There was no significant difference between the two subgroups in terms of the performed operations (p=0.140). Furthermore, there was no delay in surgery. Most of the operations were performed in an open approach (57.8%), under an elective status (91%). Additional resections were implemented in 2.4% of total cases. Additionally, a homogeneous allocation, in terms of neoadjuvant and adjuvant schemes (p=0.138), was recorded. A R0 resection was achieved in 95.5% of cases. Median follow-up for the pooled patient sample was 24 months.

Postoperative outcomes. There was no difference between older and younger patients (18.9% vs. 20%) in terms of overall complications rate (p=0.859). The pooled anastomotic leakage rate was 4.8%. Comparable readmission (p=0.67), reoperation (p=0.260) and ICU (p=0.855) admission rates were also noted. There was no difference in any other perioperative adverse event.

Primary endpoint. Regarding the primary outcome, mean survival was 26.69 months. There was no difference (28.48 vs. 25.68 months) between the two study arms (p=0.227). Similar were the results from the survival analysis function (Figure 1, Log-Rank test p=0.224).

Secondary endpoints. The optimal surgery rate was similar (p=0.074) between the younger and older subgroups (80% vs. 67%). There was no difference in terms of diagnosis to admission (p=0.973), admission to operation (p=0.893) and diagnosis to operation (p=0.992) time.

Finally, there was no survival discrepancy between the optimal and the suboptimal surgery groups (Figure 2, Log-Rank test p=0.674). Subgroup analyses (older & suboptimal, older & optimal, younger & suboptimal, younger & optimal) did not highlight any statistically significant survival difference (Figure 3, Log-Rank test p=0.467).

Colorectal cancer is among the most frequently encountered malignancies. Almost 760,000 males and 614,000 females are diagnosed with colorectal cancer every year, worldwide (19). This corresponds to almost 5% of the total population developing a colorectal tumor at least once in their life span (19).

Recent innovations in medicine have allowed an increase in the life expectancy of the Western countries’ population. It is estimated that the average life span of the developed nations ranges from 78.6 to 84.2 years, with an increasing trend (27). However, the definition of the term ‘older’ is still unclear (Geriatr Gerontol Int 6(3) (2006)">28). Conventionally, patients aged over 65 years old were regarded as older (Geriatr Gerontol Int 6(3) (2006)">28). However, in current literature, definitions up to 75 years old were reported (Geriatr Gerontol Int 6(3) (2006)">28-31). Furthermore, recent definition proposals incorporated the loss of functional independence and frailty as an additional differentiation characteristic of elderly (Geriatr Gerontol Int 6(3) (2006)">28-31). Based on previous cohorts, we defined the two subgroups of our study by the 70 years age key-point.

Overall, cancer is considered as a disease of the older patients; in terms of colorectal cancer, a parallel tendency of the prevalence rate, with the increasing age is noted (22). For instance, patients aged 60-64 years have, almost, the half incidence of colorectal malignancies, compared to the respective 80-84 years old cases. Based on these and alongside the escalation of the pooled incidence, it becomes apparent that the total number of older patients with colorectal cancer is expected to increase (22).

A multimodal approach is the current gold standard treatment for colorectal cancer (32). Radical resection combined with neo-adjuvant and adjuvant modalities allowed prolongation of the overall and disease-free survival, while in parallel minimizing the local and distant recurrence rates (32,33). However, the former, still, remains a controversial issue for older patients, since many cohorts confirmed that age is an independent prognostic factor of postoperative morbidity and mortality (21). More specifically, the increasing burden of systematic comorbidities reduces the physiological reserves, thus jeopardizing the final, postoperative outcome.

Several clinical studies have compared the two age groups in terms of perioperative endpoints. Fiscon et al. (34) reported a 24% complications rate in older patients, compared to the respective 8% of their younger counterparts. In the study of Bottino et al. (23), although no difference in overall complication rate was noted, older patients were associated with an increased risk for cardiopulmonary adverse events. In our study pooled complications rate was comparable between older and younger patients. Specific postoperative adverse event analyses showed similar results. These findings are in contrast with the results of a recent meta-analysis, where a significant difference in terms of pooled morbidity was found (20). However, Hoshino et al. (20) noted that despite statistical significance, the magnitude of difference was quite small. Possible confounders that may affect these results could be the lack of randomization, the absence of reporting standardization and the inherent heterogeneity of patient, tumor, and operation characteristics (20).

During the previous decades an age over 80 years was considered as an exclusion criterium for radical oncological resection (22). Therefore, in these patients, a palliative approach, such as diverting stoma or bypass, was often introduced. However, the advances in preoperative rehabilitation, optimization of comorbidities, intraoperative monitoring and surgical innovations allowed the safe implementation of more extensive resections (22). In our trial, oncological resections were the primarily surgical approach in both groups. Moreover, palliative procedures were performed in a comparable rate, while in parallel we confirmed that the delivery of optimal treatment was not affected by the age subgroup.

The benefits of minimal invasive colorectal resections are well established. The minimization of surgical trauma further reduces the perioperative inflammatory cascade (35). Improved cosmesis, reduced postoperative pain, enhanced mobilization and bowel function recovery are amongst the several advantages of laparoscopic colectomies (36). Nonetheless, the safety of laparoscopic colorectal resections in octogenarians is, still, a matter of controversy. The hemodynamic changes induced by the pneumoperitoneum combined with the prolonged operative times and the extreme intraoperative positions challenge the already reduced reserves of patients with cardiopulmonary comorbidities (20,21,37). However, in a clinical study by Akiyoshi et al. (37), an improved morbidity profile was suggested when a minimal invasive approach was applied in older patients. In our study 42.2% of the pooled resections were performed laparoscopically. There was no notable difference between the two groups in terms of morbidity.

Assessment of survival endpoints is a pivotal part of colorectal cancer trials. Overall and cancer-specific survival rates directly reflect the oncological efficacy of the investigated interventions. Interestingly, we could not verify any survival difference between older and younger colorectal cancer patients. Moreover, these results did not differentiate when the optimal surgery covariate was introduced. According to a systematic review by Colorectal Cancer Collaborative Group, older patients displayed a reduced overall survival, with minimal differences in cancer specific survival compared to their younger counterparts (38). Similarly, Golfinopoulos et al. reported comparable survival benefits of adjuvant and palliative chemotherapy between older and younger colorectal cancer cases (39). Moreover, colorectal cancer surgery outcomes are multifactorial and can be affected by multiple factors besides age, including stage, comorbidities, and chemotherapy scheme (38). For example, in stage IV patients, surgery has a limited effect in the overall prognosis. In our study, a relatively small cohort was assessed with a significant heterogeneity in terms of cancer staging, thus limiting the ability to safely extrapolate these results.

Our study showed that applying a suboptimal treatment in older colorectal cancer patients did not significantly affect survival outcomes. More specifically, median survival for older patients with suboptimal surgical treatment was 22.8 months compared to 25 months in the optimal arm. The best survival rate was identified in the younger and optimal subgroup (28 months). Despite these differences statistical significance was not confirmed. A possible explanation could be the previously mentioned study limitations. Current literature suggests that perioperative optimization and technological advancements allow the performance of optimal oncological resections in octogenerians with acceptable results (40). Multiple trials confirmed a survival benefit of optimal surgical treatment in older patients without documenting any safety hazard (41,42).

Despite the therapeutic potential in colorectal cancer, surgical resection represents a major physiological trauma that derails the homeostasis of a frail patient (43,44). More specifically, the extent of perioperative risk is directly correlated with the preoperative functional reserves of the patient (43,44). In octogenerians, cardiac, pulmonary, and metabolic comorbidities combined with malnutrition and other psychological disorders inhibit an early and uneventful postoperative recovery (43,44). Therefore, the introduction of multimodal pre-habilitation programs was considered as a means of improving cardiorespiratory fitness in older patients (43,44). Although these preemptive interventions typically consist of nutritional support combined with aerobic exercise and strength training, the exact protocol and duration is, still debated (43,44). Similarly, current evidence regarding the efficacy of such interventions is not, yet, conclusive (43-45). In our cohort no pre-habilitation protocol was applied.

Preoperative screening of biological vulnerability aims in the early identification of reduced reserve and stressor resistance in geriatric patients (46-48). During the previous years, multiple frailty assessment instruments were described (46-48). These were categorized in frailty phenotype instruments and frailty index tools; the former predominantly evaluated motor and activity scores, while the latter assessed comorbidities, social factors, cognition, and psychological disorders (46-48). Subsequently, a great discrepancy is noted among the results of these tools, and a gold standard frailty index is, yet, not established. Besides the plethora in assessment tools, the absence of hard evidence and implementation guidance, further prevents the systematic frailty evaluation of older patients (46-48). It is estimated that a frailty assessment algorithm is rarely introduced in daily clinical practice and that only in selected cases (49). In our study, a formal geriatric assessment was not performed. Patients were categorized based on a pre-defined age cut-off, while a validated comorbidity tool was also used.

Several limitations must be considered with regard to the present study. First, the study was designed as a prospective trial, thus increasing the risk of bias due to absence of masking and randomization algorithms. Furthermore, the relatively small sample size may have negatively affected the validity of our findings. Additionally, the differences in terms of tumor and operative characteristics were another noteworthy source of bias. Finally, divergences in the perioperative schemes, further, contribute to the overall heterogeneity.

Our cohort study highlighted that operated older colorectal cancer patients had similar survival rates with their younger counterparts. Furthermore, the two arms were comparable regarding overall morbidity and optimal surgery rates. Finally, subgroup analysis in terms of patient age and optimal surgical approach did not reveal any survival difference. However, due to several study limitations, further trials, of a higher methodological quality and of a larger sample are required to confirm these findings.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The Authors declare that they have no conflicts of interest.

Conception and design of the study: Tepetes, Aggelakopoulou; Acquisition of data: Aggelakopoulou, Mouzakis; Drafting the article: Perivoliotis, Aggelakopoulou; Critical revision: Stergiannis, Intas. Final approval: Tepetes.