Colorectal cancer (CRC) is the second most common cause of cancer-related death worldwide (1). The incidence rate of CRC considerably increases with age and age-standardized CRC mortality rates are highest among elderly patients (2, 3). Elderly metastatic CRC (mCRC) patients (≥70 years) are underrepresented in clinical trials and one out of four elderly mCRC patients does not receive chemotherapy-based palliative systemic therapy due to comorbidities, chronological age or poor performance status (4).

Therapeutic decision making and treatment recommendations by the European Society of Medical Oncology (ESMO) (5) and National Comprehensive Cancer Network (NCCN) (6) for elderly mCRC patients are mainly based on a limited number of clinical trials focusing on the elderly mCRC population (7–9). Fluorouracil-based monochemotherapy in combination with anti-VEGF-based therapy irrespective of sidedness (7, 9) or in combination with anti-EGFR-based therapy (8) as well as anti-EGFR monotherapy (10) in patients with RAS wild-type left-sided tumors represent recommended first-line protocols (5). A median overall survival of 14 and 21 months is achieved with the abovementioned first-line protocols among patients ≥75 years (10) and ≥70 years (7), respectively; however, data on the clinical outcome in the elderly mCRC population in the real-world setting are sparse.

While metastasectomy and/or local ablative treatment (+/- perioperative chemotherapy or previous conversion therapy) represent established approaches in eligible patients with oligometastatic CRC (5, 6), there is a paucity of evidence supporting this treatment concept with putative curative intent in the elderly oligometastatic CRC population.

The primary aim of this unicentric retrospective analysis was to evaluate the therapeutic management of mCRC patients ≥70 years of age and clinical outcome in a real-world setting. Furthermore, this analysis aimed at investigating the frequency, feasibility and efficacy of metastasectomy and/or local ablative treatment with putative curative intent in this elderly population.

As the aging population is highly represented among mCRC patients and due to the paucity of trial-based recommendations, therapeutic decision making in elderly mCRC patients remains challenging in clinical practice. In our unicentric, retrospective analysis we characterized patient and tumor characteristics and investigated clinical outcome in a representative elderly patient cohort undergoing systemic therapy for mCRC. The distribution of age categories within our elderly mCRC cohort was comparable to records of the Tumor Registry of the Province of Salzburg between 2013 and 2020: 70-74 years: 31% versus 32%; 75-79 years: 36% versus 35%; 80-84 years: 27% versus 33%, ≥85 years: 6% versus 0%. It is noteworthy, that only 59% of mCRC patients ≥70 years of age received palliative systemic therapy in the Province of Salzburg.

Based on the findings of our unicentric analysis we provide further evidence that OS of elderly mCRC patients undergoing systemic therapy in the real-world setting (mOS of 25.6 months) is comparable to landmark clinical trials (7, 12) (mOS of 19 to 21 months, Table A.4 ). Metastasectomy and/or local ablative treatment with curative intent proved feasible in selected elderly patients and resulted in a significant and clinically meaningful OS benefit (HR: 0.22, p<0.001, Figure 3 ). Furthermore, the observed trend towards superior OS with an anti-EGFR-based therapy in left-sided mCRC when compared to anti-VEGF-based therapy or chemotherapy alone ( Figure 2D ) sheds further light on the predictive value of sidedness and corroborates the preference of anti-EGFR-based therapy also in elderly patients with RAS/BRAF wild-type left-sided disease.

In a cross-trial comparison between our retrospective analysis and a pooled analysis (13) of the TRIBE (14) and TRIBE2 (15) study, fewer patients received subsequent therapy lines in our elderly mCRC cohort (2L: 77% versus 52%, 3L: 53% versus 27%, 4L: 27% versus 12%, 5L: 11% versus 3%, Figure 1 ). It is noteworthy that the median age at mCRC diagnosis in the aforementioned studies (TRIBE: 60.0 and 60.5 years; TRIBE2: 60.0 and 61.0 years) was considerably lower when compared to our cohort (78 years, range: 70-90 years). However, the probability to receive subsequent systemic therapy was higher in our cohort compared to mCRC patients in the AVEX trial (7) (52% versus 37%).

The chemotherapy backbone in first line (doublet or triplet versus mono) did neither impact the number of subsequent therapy lines (p=0.697, Figure 1 ), nor had an impact on clinical outcome in the entire cohort (HR: 0.83, p=0.50) or among patients not eligible for metastasectomy and/or local ablative treatment (HR: 1.11, p=0.73). The latter findings are in line with the MRC FOCUS2 (16) and FFCD 2001-02 (17) trials, where the addition of oxaliplatin (16) or irinotecan (17) to 5-FU or capecitabine did not improve OS in elderly and/or frail mCRC patients, but significantly increased the frequency of grade 3-4 toxicities (17). In this regard, it is noteworthy that the ECOG PS in our elderly cohort was comparable to the study population of the MRC FOCUS2 trial (16): ECOG 0: 21%/21%, ECOG 1: 48%/50%, ECOG 2: 26%/29%, ECOG 3: 5%/0%.

Higher treatment-related toxicity rates with a doublet chemotherapy backbone and a higher frequency of comorbidities have also been observed with increasing age in the CALGB 80405 study (18). Age demonstrated as a considerable prognostic factor in the FIRE-3 study (19) (≥65 years: 25.9 versus <65 years: 29.3 months, p=0.02) and CALGB 80405 study (18) (≥70 years versus <70 years: HR 1.32, p<0.001). Within our study population (range: 70-90 years), older patients showed a trend towards a worse ECOG PS (p=0.087), however, age as a continuous parameter did not show any additional prognostic value among mCRC patients ≥70 years ( Table 2 ).

A worse ECOG PS at mCRC diagnosis showed a statistically significant and independent association with inferior OS (2 versus 0, HR: 3.07; 3 versus 0: HR; 3.66; Figure 3 ). While classification into the ECOG PS categories (from 0: fully active to 4: completely disabled) can be rapidly performed in daily clinical practice in younger patients, the latter performance score assessment can be challenging in elderly cancer patients due to physicians’ varying conception of the usual performance spectrum of elderly people. Considerable disparities between patient-reported and physician-reported ECOG PS ratings exist (20) and there is also a poor agreement in ECOG PS ratings between clinicians (21). Other scores such as the Charlson Comorbidity Index (22), which includes age and multiple comorbidities and classifies into four risk categories, proved as predictors of survival in (m)CRC (23–25). However, our findings confirm the ECOG PS as a time-saving prognosticator and helpful tool for therapeutic decision-making (e.g. chemotherapy intensity) in daily clinical practice in elderly mCRC patients. The International Society of Geriatric Oncology recommends geriatric assessment in older cancer patients aiming at influencing treatment choice, predicting treatment-related complications and predicting clinical outcome. Geriatric assessment should include functional status, comorbidities, cognition, mental health status, fatigue, social status and support, nutrition, and the presence of geriatric syndromes (26). Based on the retrospective nature of our analysis, only the functional status was extracted from medical records and geriatric assessment was not feasible.

Contrary to the literature (27), sidedness was not prognostic among elderly mCRC patients in our cohort (left-sided versus right-sided, HR: 1.10 p=0.71, Figure 2A ), which may be explained by the application of front-line anti-VEGF-based therapy in the majority of cases with right-sided (58%) as well as left-sided (51%) primary tumor localization. Furthermore, a higher percentage of patients with right-sided primary tumors underwent metastasectomy and/or local ablative treatment (right-sided: 33% versus left-sided: 15%, Table A.3 ). This stands in contrast to the secondary metastasectomy rate among patients ≥65 years in the FIRE-3 study (19) (right-sided: 8%-13% versus left-sided: 15%-26%).

A post-hoc analysis of the FIRE-3 study in the subgroup of patients ≥ 65 years (n=199) could neither corroborate the survival benefit of cetuximab versus bevacizumab in left-sided mCRC (33.2 months versus 27.5 months, HR: 0.86, p=0.38), nor the disadvantage of first-line cetuximab-based therapy in right-sided disease (16.6 months versus 23.6 months, HR: 1.1, p=0.87) (19). Liver surgery for colorectal metastases with curative intent in elderly mCRC patients can yield a comparable OS benefit as in the young population (28, 29). For elderly mCRC patients undergoing CRC liver metastases resection an incidence of 60- to 90-day mortality ranging between 4% and 8% has been reported in population-based studies (29, 30).

According to the RAXO study, a nationwide Finnish prospective intervention study, up to 41% of mCRC patients can be classified as resectable with curative intent either upfront or after conversion therapy irrespective of chronological age (31). In our cohort, metastasectomy and/or local ablative treatment were performed in 22% of patients with technically resectable disease extent and adequate performance status and yielded a clinically meaningful and independent OS benefit (HR: 0.22, Figure 3 ). This is in line with the secondary metastasectomy rate (18%) and the OS advantage (HR: 0.44) of elderly patients in the FIRE-3 study (19). The latter findings should encourage us to identify eligible patients for metastasectomy and/or local ablative treatment with curative intent in the elderly mCRC population. The presence of liver metastases was a significant and independent negative prognostic factor (HR: 1.79, p=0.05) – presumably mainly driven by non-resectable and non-liver-limited disease.

However, we would like to emphasize that in the FIRE-3 and CALGB 80405 studies elderly patients were defined by ≥65 years and ≥70 years, respectively, and were all deemed fit for a doublet chemotherapy backbone (18, 19). Data from the Cardiovascular Health Study corroborate an increasing prevalence of frailty with higher chronological age (32). Therefore, the FIRE-3 and CALGB 80405 mCRC populations may not properly reflect the elderly and often frail mCRC population in the real-world setting.

Within the inclusion period of our retrospective analysis (2009-2022), regorafenib (33) as well as TAS-102 (34) have been established as EMA- and FDA-approved third-line therapy options based on a survival benefit versus placebo, respectively. In our cohort, one out of five patients received regorafenib and/or TAS-102 during the course of disease ( Table 1 ). Since the availability of regorafenib and TAS-102 within named patient programs or based on the respective EMA approval, our treatment strategy has not favored one drug over the other in the time interval between 2014 and 2022 ( Figure A.1 ). However, based on the toxicity profile of regorafenib (33), an increased skeletal muscle loss (35) and a higher frequency of hospitalizations with regorafenib compared to TAS-102 (36), regorafenib should be used with caution in elderly mCRC patients. Treatment with regorafenib and/or TAS-102 did not result in a survival advantage when taken into account as a time-dependent covariate (yes versus no, HR: 1.09, p=0.79, Figure 3 ). According to the ESMO Magnitude of Clinical Benefit Scale (ESMO-MCBS) (37), which is based on the extent of OS gain, QoL and toxicities, TAS-102 (MCBS: 3) proved superior to regorafenib (MCBS: 1) (5).

The SUNLIGHT study, a randomized phase 3 study comparing TAS-102 versus TAS-102 in combination with bevacizumab for third-line treatment of refractory mCRC, has met its primary endpoint, demonstrating an OS benefit with TAS-102 plus bevacizumab (10.8 months versus 7.5 months, HR: 0.61, p<0.001) (38). Due to the acceptable safety profile of TAS-102 combined with bevacizumab in previous studies (9, 38, 39) this combination may become a new third-line standard in the near future, particularly suitable for the elderly and frail mCRC population.

The availability of further new treatment options (40,) (41) within the inclusion period (2009-2022) may have also contributed to the encouraging clinical outcome (mOS of 25.6 months) compared to the experimental arm of the AVEX trial (7) (mOS of 20.7 months, Table A.4 ).

Potential limitations of our study include the retrospective nature and the length of the inclusion period (2009-2022). Within the latter time span, biomarker refinement for established therapies (42), numerous new therapies for all-comers (33, 34) and biomarker-defined targeted-therapies (40, 41) changed daily clinical practice resulting in heterogenous treatments strategies in our elderly mCRC cohort. As a consequence, the predictive biomarker status is incomplete in a relevant number of patients. Furthermore, the implementation of sidedness into first-line decision making took place after the Annual ASCO Meeting 2016 (27, 43, 44), therefore, sidedness as a predictive biomarker could only be applied in less than half of our elderly mCRC patients. It is noteworthy, that elderly patients undergoing only a best supportive care strategy were excluded from our analysis. Although the number of included patients in our analysis (n=117) was limited, the sample size was comparable to the experimental arms of the AVEX (n=140) and PANDA (n=93) landmark trials ( Table A.4 ).

Clinical outcome among real-world elderly (≥70 years) mCRC patients is comparable to the results of first-line elderly mCRC landmark trials. First-line monochemotherapy plus targeted therapy based on sidedness and molecular status should be the treatment of choice. Based on proper patient selection, one out of five elderly mCRC patients qualifies for metastasectomy and/or local ablative treatment with curative intent. A doublet chemotherapy backbone +/- targeted therapy may be expedient in elderly mCRC patients who are candidates for metastasectomy and/or local ablative treatment. The latter ablative measures are feasible and yield a clinically meaningful survival benefit in selected elderly mCRC patients.

The raw data supporting the conclusions of this article will be made available by the authors on reasonable request.

The studies involving human participants were reviewed and approved by the Ethics Committee of the provincial government of Salzburg, Austria (415-E/2343/5-2018). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

FH: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review and editing. CD: Investigation, Writing – review and editing. GR: Formal analysis, Investigation, Methodology, Writing – review and editing. KS: Investigation, Writing – review and editing. RH: Investigation, Writing – review and editing. KE: Investigation, Writing – review and editing. DN: Investigation, Writing – review and editing. EK: Investigation, Writing – review and editing. MD: Investigation, Writing – review and editing. FR: Investigation, Writing – review and editing. RG: Funding Acquisition, Resources, Supervision, Writing – review and editing. LW: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review and editing. All authors contributed to the article and approved the submitted version.