Abstract
Purpose
Although colorectal cancer incidence is declining among individuals aged over 50 years, possibly due to screening, young-onset rectal cancer (YORC) in those under 50 years of age is increasing. YORC is often diagnosed at advanced stages, leading to worse outcomes and higher mortality. This study compared short- and long-term clinical, surgical, and pathological outcomes, including survival, between patients with YORC (< 50 years) and older age groups.
Method
A retrospective single-center study of patients undergoing curative-intent resection for rectal cancer at a Norwegian university hospital from 2014 to 2024. We divided the cohort into three age groups: < 50 (YORC), 50–75, and > 75. Clinical and pathological parameters, surgical complications, and overall and disease-free survival (DFS) were compared using parametric and nonparametric analyses, Kaplan–Meier methods, and Cox proportional hazard regression.
Results
642 patients underwent curative-intent surgery for rectal cancer: YORC: n = 53 (8%); 50–75: n = 435 (68%); and > 75: n = 154 (24%). The mean age for YORC was 44 years (range 26–49). The median follow-up was 52.4 months, with an interquartile range (IQR) of 27.9–85.6 months. The oldest age group had worse DFS during follow-up (p < 0.0001, log-rank test); YORC 7 events (13.5%); 50–75 67 events (16.1%); and > 75 66 (44.9%). DFS was comparable between the YORC and the 50–75 group (p = 0.6, log-rank test). Cox models showed poorer DFS in patients aged > 75 (hazard ratio 3.04, p = 0.009) and in those with higher American Society of Anesthesiologists scores or higher pathological stages.
Conclusion
The YORC group had DFS outcomes comparable to those of patients aged 50–75 following curative-intent surgery.
What does this paper add to the literature?
While the incidence of YORC has increased over the past decades, few studies have analyzed survival beyond five years. This study compares YORC patients who were followed for a median of more than five years with groups of older patients with shorter follow-up (56.0 months for those aged 50–75 years, and 39.7 for those aged > 75). The outcomes for YORC patients, including disease-free survival, were comparable to those aged 50–75.
Keywords: Rectal cancer, Young-onset, Early-onset, Cancer-free survival, Disease-free survival, Outcomes
Introduction
Colorectal cancer (CRC) has traditionally been considered a disease of older adults, with a median age at diagnosis of 70 years. However, over the past three decades, rectal cancer epidemiology has undergone a demographic shift that demands attention from the medical community. While overall CRC incidence has declined by approximately 1% annually in individuals over 50, primarily due to successful screening programs, the incidence of young-onset rectal cancer (YORC), defined as occurring in patients under 50 years of age, has shown an alarming upward trajectory [1, 2].
The rising incidence of YORC represents a paradigm shift in our understanding of this malignancy. Recent data from the American Cancer Society indicates that individuals born in 1990 have twice the risk of developing rectal cancer at a young age compared to those born in 1950. This concerning trend has been observed across multiple geographic regions and ethnic groups, suggesting a global phenomenon [2–5].
The impact of this shift extends beyond epidemiological statistics. Young adults diagnosed with rectal cancer face unique challenges that distinguish them from their older counterparts. These individuals are often in the prime of their personal and professional lives, raising complex considerations about fertility preservation, career interruption, and long-term quality of life [6]. Furthermore, YORC is frequently diagnosed at more advanced stages, contributing to poorer outcomes and higher mortality rates among younger patients [7].
Despite this, few studies have explored the long-term outcomes of YORC. The increasing incidence of YORC necessitates a comprehensive understanding of its clinical, biological, and histological characteristics. Tailored treatment strategies, including surgical approaches and adjuvant therapies, must be developed to address the specific needs of this patient population. This study aimed to compare rectal cancer outcomes after curative-intent treatment between YORC patients and older counterparts. We hypothesized that young-onset rectal cancer was associated with poorer disease-free survival than older-onset cases, and specifically, we wanted to compare early outcomes (clinicopathological features, surgical complications, length of stay, and 30-day readmission rate) and long-term disease-free and overall survival.
Methods
This study was an analysis of a prospectively collected database of consecutive patients with histologically confirmed rectal cancer who underwent total mesorectal excision (TME) surgery with curative intent at Akershus University Hospital between 2014 and January 2024 and followed until September 2024. The database was updated using electronic health records that were automatically synchronized with the national population registry, which indicated whether patients were alive or deceased. The study was approved by the Regional Committee for Medical and Health Research Ethics (REK approval no. 832716). Passive consent from the patients was obtained in accordance with applicable Norwegian legislation. Rectal cancer was defined as any adenocarcinoma, including the mucinous subtype, within 15 cm from the anal verge measured with a rigid scope. This observational study is reported in accordance with the STROBE guidelines for cohort studies.
Guidelines for rectal cancer treatment
Patients were treated according to the Norwegian guidelines for rectal cancer [8]. All patients were included in the expedited pathway for diagnosis and treatment of CRC. After histopathological diagnosis and staging with CT of the thorax, abdomen, and pelvis and MRI of the pelvis, treatment plans were made in a multidisciplinary meeting. The main indication for neoadjuvant treatment was a short distance from the tumor to the expected circular resection margin (CRM ≤ 2 mm). Other factors considered were advanced T stage, positive N stage, and extramural vascular invasion (mrEMVI positive). Special considerations were given to tumors high and low in the rectum. Neoadjuvant radiotherapy consisted of either a short-course regimen (5 Gy × 5) or a long-course regimen (2 Gy × 25). Concomitant or sequential chemotherapy was given with neoadjuvant radiation according to the national guidelines [8]. Patients were followed routinely with clinical appointments, CT examinations, and colonoscopies up to five years after surgery. The locally most advanced rectal cancers (T4b) were referred to a tertiary cancer center and were not included in our database.
Pathology
The UICC/AJCC TNM classification was used to grade tumors [9]. A positive circumferential resection margin (CRM) was defined as a tumor distance of less than 1 mm from the resection margin. In addition, we measured the distal resection margin, the number of lymph nodes, the number of positive lymph nodes, the presence of extramural vascular invasion (EMVI), and perineural invasion. The surgical specimen was graded according to Quirke’s classification as complete, near-complete, or incomplete [10].
Statistics
We divided the patient cohort into three age groups (in years): < 50 (YORC), 50–75, and > 75. Numeric variables are described as means and standard deviations and analyzed using a one-way analysis of variance. Categorical variables are reported as numbers and percentages and compared using Pearson’s chi-square test. To avoid selection bias, we first compared < 50 versus 50–75 versus > 75 and then < 50 versus 50–75. Patients older than 75 years were excluded from these latter analyses to minimize confounding from competing mortality and treatment heterogeneity. The same comparisons were used in the survival analysis.
Dates of primary surgery, time to recurrence, sites of recurrence, and time of death were used in survival analysis. DFS is defined as the time from surgery to recurrence or death from any cause. For one patient, the surgery date was missing, and for another, the recurrence date was missing; both were excluded from the survival analysis. The Kaplan–Meier method with the log-rank test was used to estimate survival, and raincloud plots were used to display the distributions of age and follow-up time. The adjusted hazard ratio (HR) for DFS was computed using a multivariable Cox proportional hazards model. The clinically relevant variables included in this Cox model were, in addition to age groups YORC, 50–75, and > 75, sex, American Society of Anesthesiologists (ASA) score (1–2 vs. 3–4), carcinoembryonic antigen (CEA), chemotherapy, radiotherapy, Quirke’s TME score, pEMVI, and pathological cancer stage.
Patients who developed systemic metastases by the time of rectal surgery, despite treatment with curative intent, were excluded from the survival analysis. A significance level of 0.05 was used for all analyses. Reported p-values are two-tailed. R Studio 4.4.2 and Microsoft Excel were used to analyze the data.
Results
Overall, 665 patients were operated on for rectal cancer in the period from January 2014 to January 2024. Of these, 23 were excluded because they underwent palliative rectal surgery (Fig. 1).
Fig. 1.
Flow sheet
The youngest patient was 26 years old (Fig. 2). The mean age was 44 in YORC, 64 in the 50–75 group, and 80 in the > 75 group.
Fig. 2.
Age distribution of patients operated with curative intent for rectal cancer. The box plot shows the median age (middle line), the interquartile range (box) representing the middle 50% of patients, and whiskers that extend to the most extreme ages within 1.5 times the interquartile range. The youngest patient was 26 years old. Mean ages were 44, 64, and 80 in the < 50, 50–75, and > 75 groups, respectively
Body mass index was lower in the oldest age group (p = 0.018), with values of 27 kg/m2 in YORC, 26.4 kg/m2 in the 50–75 age group, and 25.3 kg/m2 in those over 75. ASA 3 score was unevenly distributed (p < 0.001): n = 0 in YORC, n = 94 (22%) in 50–75, and n = 77 (50%) in > 75. On MRI, the age groups were comparable in T stage (p = 0.6) and N stage (p = 0.4) as well as for the rates of tumor deposits, positive mrEMVI, and distance to the anal verge. Likewise, the percentage of patients with liver or lung metastases at diagnosis was similar. The mean CEA value was significantly lower in YORC compared to 50–75 and > 75 (p < 0.001) (Table 1).
Table 1.
Baseline characteristics
| Variable | All ages, n = 642 | < 50 years, n = 53 (8%) |
50–75 years, n = 435 (68%) |
> 75 years, n = 154(24%) | p1 | p2 | |
|---|---|---|---|---|---|---|---|
| Sex | Female | 244 (38%) | 26 (49%) | 160 (37%) | 58 (38%) | 0.2 | 0.11 |
| Male | 398 (62%) | 27 (51%) | 275 (63%) | 96 (62%) | |||
| Age (years) | Mean (SD) | 66 (12) | 44 (5) | 64 (7) | 80 (4) | < 0.001 | < 0.001 |
| BMI3 (kg/m2) | Mean (SD) | 26.2 (4.6) | 27.0 (4.7) | 26.4 (4.8) | 25.3 (3.7) | 0.018 | 0.4 |
| ASA classification | 1 | 48 (7.5%) | 17 (32%) | 31 (7.2%) | 0 (0%) | < 0.001 | < 0.001 |
| 2 | 416 (65%) | 36 (68%) | 305 (71%) | 75 (49%) | |||
| 3 | 171 (27%) | 0 (0%) | 94 (22%) | 77 (50%) | |||
| 4 | 2 (0.3%) | 0 (0%) | 1 (0.2%) | 1 (0.7%) | |||
| Missing4 | 5 | 0 | 4 | 1 | |||
| mrT-stage | 1 | 2 (0.3%) | 0 (0%) | 2 (0.5%) | 0 (0%) | 0.6 | 0.9 |
| 2 | 191 (30%) | 17 (33%) | 131 (30%) | 43 (28%) | |||
| 3 | 333 (52%) | 24 (46%) | 220 (51%) | 89 (58%) | |||
| 4 | 113 (18%) | 11 (21%) | 81 (19%) | 21 (14%) | |||
| Missing4 | 3 | 1 | 1 | 1 | |||
| mrN-stage | 0 | 465 (73%) | 39 (75%) | 310 (71%) | 116 (76%) | 0.4 | 0.7 |
| 1 | 128 (20%) | 8 (15%) | 89 (21%) | 31 (20%) | |||
| 2 | 46 (7.2%) | 5 (9.6%) | 35 (8.1%) | 6 (3.9%) | |||
| Missing4 | 3 | 1 | 1 | 1 | |||
| Clinical stage | I | 183 (29%) | 16 (31%) | 123 (28%) | 44 (29%) | > 0.9 | 0.9 |
| II | 274 (43%) | 23 (44%) | 183 (42%) | 68 (44%) | |||
| III | 159 (25%) | 12 (23%) | 112 (26%) | 35 (23%) | |||
| IV | 23 (3.6%) | 1 (1.9%) | 16 (3.7%) | 6 (3.9%) | |||
| Missing4 | 3 | 1 | 1 | 1 | |||
| mrTumor deposits | Positive | 13 (2.0%) | 0 (0%) | 11 (2.5%) | 2 (1.3%) | 0.4 | 0.5 |
| mrEMVI | Positive | 101 (16%) | 8 (15%) | 73 (17%) | 20 (13%) | 0.5 | > 0.9 |
| mrTumor diameter3 (cm) | Mean (SD) | 4.03 (1.63) | 3.75 (1.37) | 4.05 (1.67) | 4.05 (1.62) | 0.3 | 0.2 |
| mrAnal verge distance (cm) | Mean (SD) | 8.4 (3.5) | 8.7 (4.0) | 8.5 (3.5) | 7.9 (3.5) | 0.12 | 0.6 |
| Rectal subsite5 | Low 1–5 cm | 156 (24%) | 13 (25%) | 102 (23%) | 41 (27%) | 0.040 | 0.082 |
| Mid > 5–10 cm | 299 (47%) | 17 (32%) | 203 (47%) | 79 (51%) | |||
| High > 10 cm | 187 (29%) | 23 (43%) | 130 (30%) | 34 (22%) | |||
| mrPuborectal sling distance (cm) | Mean (SD) | 5.0 (3.5) | 5.7 (4.0) | 5.1 (3.5) | 4.6 (3.3) | 0.13 | 0.3 |
| Lung metastases | Present | 7 (1.1%) | 0 (0%) | 5 (1.1%) | 2 (1.3%) | 0.7 | > 0.9 |
| Liver metastases | Present | 18 (2.8%) | 1 (1.9%) | 13 (3.0%) | 4 (2.6%) | 0.9 | > 0.9 |
| CEA (μg/L) | Mean (SD) | 7 (17) | 3 (3) | 7 (18) | 7 (14) | < 0.001 | < 0.001 |
1Pearson’s chi-square test or one-way analysis of means (not assuming equal variance), < 50 years versus 50–75 years versus > 75 years
2Pearson’s chi-square test or one-way analysis of means (not assuming equal variance), < 50 years versus 50–75 years
3Missing for one patient
4Left out of analysis
5Measured from the anal verge
Abbreviations: BMI, body mass index; ASA classification, American Society of Anesthesiologists classification; mr, magnetic resonance imaging; EMVI, extramural vascular invasion; CEA, carcinoembryonic antigen
Among patients proceeding directly to surgery, YORC was associated with a higher rate of pT0 (p = 0.028). In these cases, cancer was identified on histological analysis following local excision, prompting rescue TME; however, no residual tumor was found in any of these TME specimens. The distributions of pN (after direct surgery) and ypN (after neoadjuvant therapy and surgery) were comparable across age groups (Table 2).
Table 2.
Oncologic outcomes
| Variable—direct surgery | All ages, n = 476 | < 50 years, n = 39 (8%) | 50–75 years, n = 315 (66%) | > 75 years, n = 122 (26%) | p1 | p2 | |
|---|---|---|---|---|---|---|---|
| pT stage | 0 | 37 (8%) | 8 (21%) | 25 (8%) | 4 (3%) | 0.028 | 0.078 |
| 1 | 49 (10%) | 4 (10%) | 31 (10%) | 14 (11%) | |||
| 2 | 150 (32%) | 13 (33%) | 90 (29%) | 47 (39%) | |||
| 3 | 216 (45%) | 13 (33%) | 151 (48%) | 52 (43%) | |||
| 4 | 24 (5%) | 1 (3%) | 18 (6%) | 5 (4%) | |||
| pN stage | 0 | 320 (67%) | 30 (77%) | 205 (65%) | 85 (70%) | 0.4 | 0.2 |
| 1 | 115 (24%) | 5 (13%) | 84 (27%) | 26 (21%) | |||
| 2 | 41 (9%) | 4 (10%) | 26 (8%) | 11 (9%) | |||
| Total lymph nodes (no.) | Mean (SD) | 26 (11) | 29 (11) | 27 (12) | 24 (10) | 0.008 | 0.2 |
| Positive lymph nodes (no.) | Mean (SD) | 1.00 (2.44) | 0.92 (2.28) | 1.01 (2.31) | 1.02 (2.81) | > 0.9 | 0.8 |
| pEMVI | Positive | 147 (31%) | 11 (28%) | 96 (30%) | 40 (33%) | 0.8 | > 0.9 |
| pPn | Positive | 74 (16%) | 4 (10%) | 51 (16%) | 19 (16%) | 0.6 | 0.5 |
| Differentiation | High | 118 (27%) | 7 (23%) | 78 (27%) | 33 (28%) | 0.8 | > 0.9 |
| Middle | 255 (58%) | 18 (60%) | 169 (58%) | 68 (58%) | |||
| Low | 40 (9%) | 3 (10%) | 29 (10%) | 8 (7%) | |||
| Mucinous | 24 (6%) | 2 (7%) | 13 (5%) | 9 (8%) | |||
| Missing3 | 39 | 9 | 26 | 4 | |||
| CRM4 (mm) | Mean (SD) | 10 (9) | 14 (11) | 10 (10) | 10 (8) | 0.2 | 0.07 |
| DRM5 (mm) | Mean (SD) | 32 (17) | 37 (23) | 31 (16) | 31 (18) | 0.3 | 0.15 |
| Adjuvant radiotherapy | Yes | 14 (3%) | 2 (5%) | 11 (4%) | 1 (1%) | 0.2 | > 0.9 |
| Adjuvant chemotherapy6 | Yes | 77 (16%) | 9 (23%) | 65 (21%) | 3 (3%) | < 0.001 | 0.9 |
| Variable—neoadjuvant therapy and surgery | All ages, n = 166 | < 50 years, n = 14 (8%) | 50–75 years, n = 120 (72%) | > 75 years, n = 32 (19%) | p1 | p2 | |
| Therapy | Radiation2 Gy × 25 | 85 (51%) | 10 (71%) | 66 (55%) | 9 (28%) | 0.007 | 0.4 |
| Radiation 5 Gy × 5 | 81 (49%) | 4 (29%) | 54 (45%) | 23 (72%) | 0.007 | 0.4 | |
|
Chemo- therapy |
113 (68%) | 12 (86%) | 90 (75%) | 11 (35%) | < 0.001 | 0.6 | |
| ypT stage | 0 | 31 (19%) | 4 (29%) | 23 (19%) | 4 (13%) | 0.7 | 0.6 |
| 1 | 10 (6%) | 2 (14%) | 7 (6%) | 1 (3%) | |||
| 2 | 52 (31%) | 3 (21%) | 39 (33%) | 10 (31%) | |||
| 3 | 71 (43%) | 5 (36%) | 49 (41%) | 17 (53%) | |||
| 4 | 2 (1%) | 0 (0%) | 2 (2%) | 0 (0%) | |||
| ypN stage | 0 | 104 (63%) | 6 (43%) | 78 (65%) | 20 (63%) | 0.2 | 0.5 |
| 1 | 43 (26%) | 4 (29%) | 32 (27%) | 7 (22%) | |||
| 2 | 19 (11%) | 4 (29%) | 10 (8%) | 5 (16%) | |||
| ypCR | Yes | 27 (16%) | 2 (14%) | 22 (18%) | 3 (9%) | 0.5 | > 0.9 |
| Total lymph nodes (no.) | Mean (SD) | 19 (9) | 24 (11) | 18 (9) | 20 (9) | 0.1 | 0.07 |
| Positive lymph nodes (no.) | Mean (SD) | 1.06 (1.98) | 2.00 (3.09) | 0.88 (1.64) | 1.31 (2.46) | 0.3 | 0.2 |
| ypEMVI | Positive | 35 (21%) | 2 (14%) | 27 (23%) | 6 (19%) | 0.7 | 0.7 |
| ypPn | Positive | 28 (17%) | 1 (7%) | 23 (19%) | 4 (13%) | 0.4 | 0.5 |
| Differentiation | High | 36 (22%) | 1 (7%) | 28 (23%) | 7 (22%) | 0.4 | 0.2 |
| Middle | 54 (33%) | 5 (36%) | 36 (30%) | 13 (41%) | |||
| Low | 6 (4%) | 2 (14%) | 4 (3%) | 0 (0%) | |||
| Mucinous | 6 (4%) | 0 (0%) | 5 (4%) | 1 (3%) | |||
| Unclassifiable | 64 (39%) | 6 (43%) | 47 (39%) | 11 (34%) | |||
| CRM7 (mm) | Mean (SD) | 6.7 (8.9) | 4.2 (3.7) | 7.0 (9.1) | 6.6 (9.8) | 0.2 | 0.06 |
| DRM8 (mm) | Mean (SD) | 34 (17) | 35 (16) | 34 (17) | 36 (18) | 0.8 | 0.8 |
| Adjuvant chemotherapy9 | Yes | 11 (7%) | 1 (7%) | 8 (7%) | 2 (6%) | > 0.9 | > 0.9 |
1Pearson’s chi-square test or one-way analysis of means (not assuming equal variance), < 50 years versus 50–75 years versus > 75 years
2Pearson’s chi-square test or one-way analysis of means (not assuming equal variance), < 50 years versus 50–75 years
3Left out of the analysis
4Missing for 34 patients
5Missing for 27 patients
6For patients not receiving neoadjuvant treatment
7Missing for 30 patients
8Missing for 32 patients
9For patients receiving neoadjuvant treatment
Abbreviations: CR, complete response; CRM, circumferential resection margin; DRM, distal resection margin; EMVI, extramural vascular invasion; p, histopathologic status after direct surgery; Pn, perineural infiltration; yp, histopathologic status after neoadjuvant therapy and surgery
Neoadjuvant and adjuvant treatment
Among patients receiving neoadjuvant treatment, the radiation modality used differed as expected across age groups. The > 75 group received short-course radiation (5 Gy × 5) more often than the younger patients (p = 0.007) and had a lower probability of receiving chemotherapy (p < 0.001). A higher proportion of younger patients received adjuvant chemotherapy (p < 0.001) (Table 2).
Postoperative pathological status
The total number of lymph nodes retrieved in specimens after direct surgery was higher among young patients (p = 0.008), with a mean of 29 (SD 11) in the YORC group, a mean of 27 (SD 12) in the 50–75 age group, and a mean of 24 (SD 10) in the > 75 age group. For patients receiving neoadjuvant therapy, no such difference was observed (p = 0.1). The number of positive lymph nodes was also similar across the age groups (Table 2).
Surgical treatment
More young patients underwent transanal total mesorectal excision (TaTME; p < 0.018), specifically, 17% of YORC, 6.7% of those aged 50–75, and 11% of those aged > 75. The rate of Sigmoid stoma increased with age (p < 0.001): 19% in YORC, 32% in 50–75, and 68% in < 75. By contrast, more young patients underwent primary anastomosis (p < 0.001). When comparing the YORC and 50–75 groups, the rates of stoma and anastomosis were comparable (p = 0.068 and p = 0.13). The rates of anastomotic leakage and endosponge use were similar across all age groups, as was the quality of the pathological specimen, as assessed by Quirke’s score. The length of stay, 30-day readmission rate, and complication rate (Clavien–Dindo grade ≥ 3) were not significantly different between the groups (Table 3).
Table 3.
Surgical treatment, complications, and short-term outcomes
| Variable | All ages, n = 642 | < 50 n = 53 (8%) |
50–75 n = 435 (68%) |
> 75 n = 154 (24%) |
p1 | p2 |
|---|---|---|---|---|---|---|
| Surgical approach | ||||||
| Laparoscopic | 345 (54%) | 24 (45%) | 242 (56%) | 79 (51%) | 0.3 | 0.2 |
| Open | 140 (22%) | 11 (21%) | 92 (21%) | 37 (24%) | 0.7 | > 0.9 |
| Robot | 102 (16%) | 9 (17%) | 72 (17%) | 21 (14%) | 0.7 | > 0.9 |
| TaTME | 55 (8.6%) | 9 (17%) | 29 (6.7%) | 17 (11%) | 0.018 | 0.018 |
| Converted | 50 (7.8%) | 5 (9.4%) | 36 (8.3%) | 9 (5.8%) | 0.6 | > 0.9 |
| Low anterior | 383 (60%) | 34 (64%) | 256 (59%) | 93 (60%) | 0.7 | 0.6 |
| High anterior | 117 (18%) | 11 (21%) | 83 (19%) | 23 (15%) | 0.5 | > 0.9 |
| APR | 142 (22%) | 8 (15%) | 96 (22%) | 38 (25%) | 0.3 | 0.3 |
| Sigmoid stoma | 254 (40%) | 10 (19%) | 140 (32%) | 104 (68%) | < 0.001 | 0.068 |
| Anastomosis | 373 (58%) | 41 (77%) | 287 (66%) | 45 (29%) | < 0.001 | 0.13 |
| Anastomotic leakage3 | 61 (9.5%) | 8(15%) | 45 (10%) | 8 (5.2%) | 0.061 | 0.4 |
| Reoperation for leakage4 | 19 (3.0%) | 3 (5.7%) | 14 (3.2%) | 2 (1.3%) | 0.2 | 0.6 |
| Endosponge | 33 (5.1%) | 2 (3.8%) | 25 (5.7%) | 6 (3.9%) | 0.6 | 0.8 |
| Abcess/sinus5 | 27 (4.2%) | 2 (3.8%) | 19 (4.4%) | 6 (3.9%) | > 0.9 | > 0.9 |
| Clavien–Dindo ≥ 3 | 133 (21%) | 8 (15%) | 95 (22%) | 30 (19%) | 0.5 | 0.3 |
| Length of stay (days)6 | 8.4 (6.4) | 7.9 (6.0) | 8.2 (6.6) | 8.9 (6.1) | 0.4 | 0.7 |
| Readmission within 30 days | 127 (20%) | 6 (11%) | 88 (20%) | 33 (21%) | 0.3 | 0.2 |
| Quirckes TME score | ||||||
| Complete | 437 (73%) | 39 (81%) | 297 (73%) | 101 (68%) | 0.11 | 0.3 |
| Near complete | 84 (14%) | 6 (13%) | 49 (12%) | 29 (19%) | ||
| Incomplete | 81 (13%) | 3 (6.3%) | 59 (15%) | 19 (13%) | ||
| Missing7 | 40 | 5 | 30 | 5 | ||
1Pearson’s chi-square test < 50 years vs. 50–75 years vs. > 75 years
2Pearson’s chi-square test < 50 years vs. 50–75 years
3Including sinus and abscess for patients with anastomosis
4Missing for one patient
5For patients without anastomosis
6Mean (SD)
7Left out of the analysis
Abbreviations: TaTME, transanal total mesorectal excision; APR, abdominoperineal resection; TME, total mesenteric excision
Disease-free and overall survival
The median follow-up time for the entire cohort was 52.4 months (IQR, 27.9–85.6); 68.5 months (IQR, 30.0–96.1) for YORC; 56.0 months (IQR, 28.6–90.0) for 50–75; and 39.7 months (IQR, 22.8–69.5) for those aged > 75 (Figs. 3 and 4). The oldest patients had inferior DFS (p < 0.0001, log-rank test). However, no difference was found between YORC and 50–75 (p = 0.6, log-rank test) (Fig. 5). We found a similar pattern for overall survival: poorer in the > 75 group (p < 0.0001, log-rank test), whereas no difference was observed between the YORC and 50–75 groups (p = 0.39, log-rank test) (Fig. 6). In the Cox proportional hazards model, the age group > 75 years was a significant predictor of impaired DFS (HR 3.04, p = 0.009). Also, in the Cox model, no difference was found between YORC and 50–75 (HR 0.92, p = 0.83). As would be expected, ASA 3–4 (versus ASA 1–2) was associated with an increased risk of a DFS event (HR 1.62, p = 0.013), as were pathological stages 2 and 3 (versus stage 0; HR 3.67, p = 0.035 and HR 6.15, p = 0.003) (Table 4).
Fig. 3.
Raincloud plots showing follow-up time (in months) by age group: < 50, 50–75, and > 75 years. Each plot includes a box plot (showing the median and interquartile range) and a density curve to illustrate the distribution
Fig. 4.
Flow sheet of patients included in disease-free survival analysis
Fig. 5.
Disease-free survival by age groups as indicated. If excluding patients > 75 years: Log-rank test p = 0.6
Fig. 6.
Overall survival by age groups as indicated. If excluding patients > 75 years: Log-rank test p = 0.39
Table 4.
Multivariable Cox proportional hazards analysis of disease-free survival
| Variable | HR | 95% CI (lower) | 95% CI (upper) | p-value |
|---|---|---|---|---|
| Age 50–751 | 0.92 | 0.41 | 2.04 | 0.83 |
| Age > 751 | 3.04 | 1.33 | 6.94 | 0.009 |
| Male vs. female sex | 1.21 | 0.84 | 1.74 | 0.31 |
| ASA 3–4 vs. ASA 1–2 | 1.62 | 1.11 | 2.36 | 0.013 |
| CEA | 1.01 | 1.00 | 1.02 | 0.21 |
| Chemotherapy | 1.46 | 0.67 | 3.19 | 0.34 |
| Radiation therapy | 0.78 | 0.41 | 1.46 | 0.43 |
| Near complete TME2 | 1.15 | 0.71 | 1.86 | 0.58 |
| Incomplete TME2 | 1.08 | 0.65 | 1.80 | 0.76 |
| pEMVI pos vs. neg | 1.20 | 0.83 | 1.74 | 0.34 |
| pStage 13 | 1.92 | 0.57 | 6.44 | 0.29 |
| pStage 23 | 3.67 | 1.10 | 12.3 | 0.035 |
| pStage 33 | 6.15 | 1.89 | 20.0 | 0.003 |
1Compared with < 50 years
2Compared with patients who underwent complete TME
3Compared with pStage 0
Abbreviations: pEMVI, pathologic extramural venous invasion; CEA, carcinoembryonic antigen; TME, total mesenteric excision
CI lower and upper denotes the 95% confidence interval
Discussion
In our cohort of patients operated for rectal cancer, we compared the outcomes between age groups. There was no significant difference in short-term outcomes, including complications such as anastomotic leaks, abscesses, or Clavien–Dindo ≥ 3. The length of stay and readmission rates were also similar. As expected, the oldest age group (> 75) had poorer overall survival and DFS. But there was no significant difference in OS or DFS between the YORC group and the 50–75 age group.
Consistent with current assumptions, this study showed that younger rectal cancer patients receive more neoadjuvant and adjuvant treatment than older patients. However, the differences between YORC and those aged 50–75 years were not statistically significant. Interestingly, more lymph nodes were retrieved in the YORC group, but only in patients who had not received neoadjuvant treatment. In contrast, the number of positive lymph nodes was similar across all age groups. The surgical treatment was comparable, except that a larger proportion of the YORC group underwent TaTME, fewer young patients received sigmoid stomas, and more young patients underwent anastomosis. It is reasonable to believe that surgeons are reluctant to perform primary anastomosis in older patients because of frailty and the more severe consequences of anastomosis leakage [11]. For Sigmoid stoma and anastomosis, the YORC and 50–75 groups had similar outcomes.
According to Quirke’s score, pathological specimen quality was excellent across all age groups. There were no differences between age groups in rates of anastomotic leakage, major complications (Clavien–Dindo ≥ 3), length of hospital stay, or 30-day readmission rates. The YORC patients and those aged 50–75 had similar OS and DFS, corroborated by the Cox proportional hazards model. The oldest age group had an increased risk of a DFS event and worse overall survival.
In general, there is an epidemiological shift and an increase in the incidence of young-onset CRC [7, 12]. This is in line with our study findings, in which 8% were YORC, including several individuals in their twenties. This epidemiological shift has been demonstrated in several comprehensive population-based studies focusing on CRC in younger age groups [7, 13, 14]. Vuik and colleagues reported in 2019, in a comprehensive population-based study spanning 20 European countries over 25 years, an alarming rise in CRC among patients aged 20–29 years, with an annual incidence increase of 7–8% [7]. The biological basis for this increasing incidence still needs to be understood. While hereditary syndromes such as Lynch syndrome and familial adenomatous polyposis account for approximately 20% of young cases, the majority arise as sporadic CRC in individuals without a known genetic predisposition [12].
YORC has distinct features that differ from young-onset CRC. Lifestyle variables, including food, exercise, and consumption of alcohol, are related to the risk of cancer in the colon but not to rectal cancer. Similarly, there is varying evidence on the impact of obesity on colon and rectal cancer, indicating that obesity is more specifically linked to colon cancer [15, 16]. The association of obesity with rectal cancer is less evident, and of note, in our series, the mean body mass index was well below the obesity threshold.
This study showed that younger rectal cancer patients received more radiation and chemotherapy in the neoadjuvant setting, which aligns with findings from Zaborowski et al. [13, 17]. However, we found no such differences between the younger-onset rectal cancer (YORC) group and the 50–75 age group. Notably, most of our patients < 75 years old had ASA class 1 or 2 scores, which may impact the decision to pursue neoadjuvant treatment. Adjuvant chemotherapy was also administered more often in the younger age groups.
Our study was a retrospective, single-center study with inherent limitations.
First, treatment guidelines vary internationally, which may introduce differences when comparing our results with those reported by others. Second, to compare the younger rectal cancer patients with their older counterparts, we divided patients into three age groups. We dichotomized the data when comparing the YORC and 50–75 groups. Dichotomization is prone to loss of statistical power and to both type I and type II errors; therefore, the results must be interpreted with caution. Third, we lacked access to patient-centered functional outcomes.
YORC is frequently diagnosed at more advanced stages, with poorer outcomes than for average-onset rectal cancer [7]. A limitation of our study is that the most locally advanced cancers (T4b) were not included in our analyses, as these patients are routinely referred and treated at a tertiary cancer center specializing in rectal cancer treatment beyond TME. Also, patients who were not treated with TME surgery, for instance, those receiving only palliative oncologic treatment, were not included. As earlier studies have indicated, younger patients present with more advanced cancer (7).
This could have introduced skewness in our data.
Additionally, the lower perioperative CEA in the YORC cohort raises concerns about potential bias in our data. The YORC patients may have had less advanced disease than older patients due to a selection bias. The lower number of lymph nodes retrieved in the older cohort could reflect less radical surgery or an unintentionally limited lymph node search by the pathologist. An alternative hypothesis is that older patients are less immunogenic than younger patients and therefore provide fewer viable lymph nodes on histopathological examinations. Given these potential biases and the fact that older patients received less intensive neoadjuvant treatment, one should be cautious in concluding that the absence of significant differences in ypT, ypN, ypCR, or positive lymph nodes implies similar responses to neoadjuvant treatment across age groups.
The study’s strengths include the large number of patients surgically treated for rectal cancer, prospectively collected data, and a follow-up exceeding 5 years in the YORC group.
In conclusion, YORC patients operated with curative intent had comparable DFS and overall survival with those aged 50–75 years, with no apparent differences in the pathological stages at the time of surgery. We have demonstrated that if a curative-intent treatment course is feasible, young patients can achieve excellent outcomes.
Acknowledgements
The authors would like to thank the patients who participated in this study and the healthcare professionals involved in rectal cancer care at Akershus University Hospital.
Author contribution
O.F.J., R.R., and K.M.A. curated the data and wrote the first draft of the manuscript. O.F.J., R.R., S.M., A.H.R., and K.M.A. contributed to conceptualization and approved the final manuscript. O.F.J. and K.M.A. prepared the tables, and O.F.J. prepared the figures. All authors reviewed the manuscript.
Funding
Open access funding provided by Akershus University Hospital (AHUS) This work was supported by The Norwegian Cancer Society—Grant 279490-2024.
Data availability
The data used in this study come from an internal quality register and are protected under Norwegian law. Due to privacy regulations, the data cannot be publicly shared. Access may be granted upon request, subject to ethical approval and compliance with applicable laws.
Declarations
Ethics approval
The study is approved by the Regional Committee for Medical and Health Research Ethics (REK approval no. 832716). The hospital data protection officer approved the study protocol (PVO reference number 2024_87).
Consent to participate
Passive consent was obtained in accordance with applicable Norwegian legislation.
Conflict of interest
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Oddne Fjellstad Johnsen and Rolf Nerem Riis shared the first authorship and contributed equally to writing the article.
References
- 1.Akimoto N, Ugai T, Zhong R, Hamada T, Fujiyoshi K, Giannakis M et al (2021) Rising incidence of early-onset colorectal cancer - a call to action. Nat Rev Clin Oncol 18(4):230–243 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Siegel RL, Miller KD, Goding Sauer A, Fedewa SA, Butterly LF, Anderson JC et al (2020) Colorectal cancer statistics, 2020. CA Cancer J Clin 70(3):145–164 [DOI] [PubMed] [Google Scholar]
- 3.Ren B, Yang Y, Lv Y, Liu K (2024) Survival outcome and prognostic factors for early-onset and late-onset metastatic colorectal cancer: a population based study from SEER database. Sci Rep 14(1):4377 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Montminy EM, Zhou M, Maniscalco L, Penrose H, Yen T, Patel SG et al (2021) Trends in the incidence of early-onset colorectal adenocarcinoma among Black and White US residents aged 40 to 49 years, 2000–2017. JAMA Netw Open 4(11):e2130433 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tsai MH, Moore JX, Cabral DN, Rogers CR, Cortes J (2024) Temporal trends in early-onset colorectal cancer incidence (2000–2020) by age group and five geographic regions in the state of Georgia. Am J Cancer Res 14(1):182–191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Waddell O, McLauchlan J, McCombie A, Glyn T, Frizelle F (2023) Quality of life in early-onset colorectal cancer patients: systematic review. BJS Open 7(3). 10.1093/bjsopen/zrad030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Vuik FE, Nieuwenburg SA, Bardou M, Lansdorp-Vogelaar I, Dinis-Ribeiro M, Bento MJ et al (2019) Increasing incidence of colorectal cancer in young adults in Europe over the last 25 years. Gut 68(10):1820–1826 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wibe A ea. Kreft i tykktarm og endetarm: Handlingsprogram: Helsedirektoratet; first published 2010 [updated 2024 January 31]. Available from: https://www.helsedirektoratet.no/retningslinjer/kreft-i-tykktarm-og-endetarm-handlingsprogram. Accessed 29 Apr 2025
- 9.Brierley JD, Gospodarowicz MK, Wittekind C (2017) TNM classification of malignant tumours. Wiley [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Glimelius B, Beets-Tan R, Blomqvist L, Brown G, Nagtegaal I, Pahlman L et al (2011) Mesorectal fascia instead of circumferential resection margin in preoperative staging of rectal cancer. J Clin Oncol 29(16):2142–2143 [DOI] [PubMed] [Google Scholar]
- 11.Mariusdottir E, Jorgren F, Saeed M, Wikstrom J, Lydrup ML, Buchwald P (2024) Hartmann’s procedure in rectal cancer surgery is often an intraoperative decision: a retrospective multicenter study. Langenbecks Arch Surg 409(1):55 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mauri G, Patelli G, Sartore-Bianchi A, Abrignani S, Bodega B, Marsoni S et al (2024) Early-onset cancers: biological bases and clinical implications. Cell Rep Med 5(9):101737. 10.1016/j.xcrm.2024.101737 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Collaborative R, Zaborowski AM, Abdile A, Adamina M, Aigner F, d’Allens L et al (2021) Characteristics of early-onset vs late-onset colorectal cancer: a review. JAMA Surg 156(9):865–874 [DOI] [PubMed] [Google Scholar]
- 14.Patel SG, Karlitz JJ, Yen T, Lieu CH, Boland CR (2022) The rising tide of early-onset colorectal cancer: a comprehensive review of epidemiology, clinical features, biology, risk factors, prevention, and early detection. Lancet Gastroenterol Hepatol 7(3):262–274 [DOI] [PubMed] [Google Scholar]
- 15.Wei EK, Giovannucci E, Wu K, Rosner B, Fuchs CS, Willett WC et al (2004) Comparison of risk factors for colon and rectal cancer. Int J Cancer 108(3):433–442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Saeed U, Myklebust TÅ, Robsahm TE, Kielland MF, Møller B, Skålhegg BS et al (2023) Risk and survival in colorectal cancer with increasing body mass index: a nationwide population-based cohort study. Colorectal Dis 25(3):375–385 [DOI] [PubMed] [Google Scholar]
- 17.Zaborowski AM, Murphy B, Creavin B, Rogers AC, Kennelly R, Hanly A et al (2020) Clinicopathological features and oncological outcomes of patients with young-onset rectal cancer. Br J Surg 107(5):606–612 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used in this study come from an internal quality register and are protected under Norwegian law. Due to privacy regulations, the data cannot be publicly shared. Access may be granted upon request, subject to ethical approval and compliance with applicable laws.