Abstract
Background:
The incidence of colorectal cancer in patients <50 years has rapidly risen recently. Understanding the presenting symptoms may facilitate earlier diagnosis. We aimed to delineate patient characteristics, symptomatology, and tumor characteristics of colorectal cancer in a young population.
Methods:
A retrospective cohort study was conducted evaluating patients <50 years diagnosed between 2005–2019 with primary colorectal cancer at a university teaching hospital. The number and character of colorectal cancer-related symptoms at presentation was the primary outcome measured. Patient and tumor characteristics were also collected.
Results:
Included were 286 patients with median age of 44 years, with 56% <45 years. Nearly all patients (95%) were symptomatic at presentation with 85% having two or more symptoms. The most common symptoms were pain (63%), followed by change in stool habits (54%), rectal bleeding (53%), and weight loss (32%). Diarrhea was more common than constipation. More than 50% had symptoms for at least three months prior to diagnosis. The number and duration of symptoms were similar in patients older than 45 compared to those younger. Most cancers were left-sided (77%) and advanced stage at presentation (36% stage III, 39% stage IV).
Conclusions:
In this cohort of young patients with colorectal cancer, the majority presented with multiple symptoms having a median duration of three months. It is essential that providers be mindful of the ever-increasing incidence of colorectal malignancy in young patients, and that those with multiple, durable symptoms should be offered screening for colorectal neoplasms based on symptoms alone.
Keywords: Colorectal cancer, Symptomatology, Young, Diagnosis
This is a retrospective study at a single referral center of 286 young patients diagnosed with colorectal cancer which aims to delineate their presenting symptoms and time to diagnosis. Understanding this pattern is essential to caregivers offering diagnostic testing given the rising incidence of colorectal cancer in this population.
Introduction
Colorectal cancer (CRC) is increasing in younger populations while declining in adults 55 years and older.(1–10) The rate of patients under 50 years in the United States being diagnosed with CRC increased from 6% in 1990 to 11% in 2013, with 72% of these occurring in patients under the age of 40.(3) Similar trends have been observed throughout the world.(11) Younger patients more often present with advanced disease and metastases.(12, 13)
In younger CRC patients, single-institutional studies demonstrate a predominance of left-sided cancer.(6, 14–17) Timely cancer detection in these patients remains elusive, and the majority of young CRC patients present in stages III-IV.(18) Reasons for advanced stage at diagnosis include lack of screening colonoscopies in young patients, vague symptoms that are explained by common and benign maladies, and higher rates of under-insurance and limited access to medical care.(12, 19, 20)
Information on clinical presentation could aid cancer recognition in this vulnerable population. While the trend towards increasing CRC diagnoses in young adults is now well-documented, data on their clinical presentation are more sparse. National databases often do not capture symptom quality and duration, making detailed, retrospective chart reviews valuable in addressing this question. One single institution study by Dozois et al. evaluated 1025 young patients without predisposing factors for CRC and found frequent symptoms were rectal bleeding, abdominal pain, and change in stool habits.(21) However, this study spans 26 years and includes patients prior to the noted rise in young-onset CRC, making additional quality data necessary.
The aims of this study were to examine contemporary patients under age 50 with primary CRC at a tertiary care center and to evaluate presenting symptom quality, frequency, and duration. Secondary aims were to explore symptom associations with patient demographics, tumor characteristics, and survival.
Materials and Methods
All patients <50 years at time of CRC diagnosis between 2005–2019 were identified via a retrospectively maintained database. The study was conducted at the University of Iowa, a National Cancer Institute-designated comprehensive cancer center. Patients were excluded if they had neuroendocrine histology or carcinoma in situ. This study was approved by the University of Iowa Institutional Review Board.
Data collected included demographics, presenting symptomatology, cancer and inflammatory bowel disease history, family history, diagnostic testing performed, tumor and germline genetic information, tumor characteristics, treatment, and survival outcomes. Demographics included age at presentation, sex, patient-defined race, body mass index (BMI) and insurance status. BMI was recorded at the time of presentation or at closest appointment. Family history of cancer was defined as any reported first, second, or third degree relative. All reported symptoms were recorded from patients’ initial time of presentation. Symptomatology was categorized as rectal bleeding, abdominal pain (including pressure), rectal pain (including pressure), back pain, bloating or distention, weight loss, fatigue or malaise, fever, anorexia, obstruction, anemia, upper gastrointestinal symptoms, constipation, diarrhea, change in stool habits, “hemorrhoids,” and urinary symptoms. Upper gastrointestinal symptoms were defined to include heartburn, reflux, indigestion, and dysphagia. “Hemorrhoids” was noted when patient attributed symptoms to “hemorrhoids.” Urinary symptoms were defined to include urgency, frequency, retention, or known urinary tract infections. Duration of symptoms prior to diagnosis, when available, was obtained. Tumor characteristics recorded included site of the tumor, tumor histology, clinical and pathological stage, molecular mutations such as microsatellite status and RAS/RAF mutations. Treatments were recorded. Outcomes collected included follow-up time-period, alive/deceased status, and progression/recurrence of disease. Follow up time was last documented visit or date of death.
Descriptive analysis evaluated trends in symptomatology and duration of symptoms. Chi-squared, Fisher’s exact test, and one-way analysis of variation (ANOVA) were used to compare presence of symptoms and the quantity and duration of symptoms in subgroups. Left/right tumor locations were compared. Left-sided cancers included splenic flexure, descending colon, sigmoid colon, and rectum as data regarding the location along the transverse colon for these tumors were not available. Age cutoff of 45 years was chosen for age group comparisons due to the recent change in colonoscopy screening guidelines. All tests were two sided with significance of p<0.05. The false-discovery rate correction was used to adjust for multiple comparisons when appropriate.(22) Overall survival since date of presentation was the primary survival endpoint with recurrence/progression-free survival as a secondary endpoint (23, 24). Cox models tested association of presenting symptoms with these outcomes with log-rank test for significance.
Results
Patient Characteristics
A total of 286 patients were diagnosed with CRC between 2005 and 2019 (Table 1). The majority of patients were males (n=150, 52.4%) and non-Hispanic white (n=248, 86.7%). Median age was 44.0 years (range 17 to 49 years), with 55.6% of patients younger than 45 years and 26.2% younger than 40 years. Twenty-one patients had a personal history of cancer; the most common included renal cancer, breast cancer, testicular cancer, gynecologic cancers, and leukemia/lymphoma. There were 46 patients with a diagnosis of inflammatory bowel disease, with the majority being ulcerative colitis (41 patients, 89.1%). Family history was positive for cancer in 181 patients (62.8%) with most common cancers being colon or rectal cancer, breast cancer, and lung cancer. Forty patients reported a first-degree family member with CRC. Seven patients had a family history of familial adenomatous polyposis or hereditary nonpolyposis CRC.
Table 1:
Patient Characteristics
| Demographics | n (%) |
|---|---|
| Age at diagnosis | |
| 17–24 | 7 (2.4) |
| 25–29 | 7 (2.4) |
| 30–34 | 17 (5.9) |
| 35–39 | 44 (15.4) |
| 40–44 | 84 (29.4) |
| 45–49 | 127 (44.4) |
| Sex | |
| Female | 136 (47.6) |
| Male | 150 (52.4) |
| Race | |
| Non-Hispanic White | 248 (86.7) |
| Non-Hispanic Black | 19 (6.6) |
| Hispanic | 12 (4.2) |
| Asian | 4 (1.4) |
| Other | 3 (1.0) |
| Body Mass Index (BMI) | |
| ≤18.5 | 12 (4.2) |
| 18.5–24.9 | 83 (29.0) |
| 25–29.9 | 70 (24.5) |
| 30–34.9 | 56 (19.6) |
| ≥35 | 60 (20.9) |
| Not available | 5 (1.7) |
| Inflammatory Bowel Disease (IBD) | |
| Ulcerative Colitis | 41 (14.3) |
| Crohn’s | 5 (1.7) |
| Personal Cancer History | |
| Renal | 4 (19.1) |
| Breast | 3 (14.3) |
| Gynecologic: ovarian, endometrial, cervical | 3 (14.3) |
| Leukemia, lymphoma | 3 (14.3) |
| Testicular | 3 (14.2) |
| Melanoma | 2 (9.5) |
| Skin (squamous cell, basal cell) | 2 (9.5) |
| Multiple cancers | 2 (9.5) |
| Gastric | 1 (4.8) |
| Craniopharyngioma | 1 (4.8) |
| Endometroid sarcoma | 1 (4.8) |
| Family History | |
| Colon and rectal cancer | 114 (39.9) |
| Breast cancer | 37 (12.9) |
| Lung cancer | 33 (11.5) |
| Prostate cancer | 22 (7.7) |
| Uterine, ovarian, or cervical cancer | 17 (5.9) |
| Lymphoma, leukemia | 13 (4.5) |
| Upper gastrointestinal: esophageal, gastric, | 12 (4.2) |
| duodenal cancer | |
| Pancreatic cancer | 12 (4.2) |
| Brain cancer | 11 (3.8) |
| Bladder/kidney cancer | 9 (3.1) |
| FAPa/HNPCCb | 7 (2.4) |
| Skin cancer/melanoma | 6 (2.1) |
| Hepatobiliary cancer | 5 (1.7) |
| Thyroid cancer | 2 (0.7) |
FAP: Familial adenomatous polyposis
HNPCC: Hereditary non-polyposis colon cancer
Tumor Characteristics
The primary diagnosis was rectal cancer in 133 patients and colon cancer in 153 patients (Table 2). More patients had left-sided cancer (n=218, 76.8%) compared to right-sided cancer (n=66, 23.2%). Two colon cancers without specified site were excluded from left/right comparison analysis.
Table 2:
Tumor Characteristics
| n (%) | |
|---|---|
| Primary Tumor Location | |
| Cecum | 28 (9.7) |
| Ascending Colon | 17 (5.9) |
| Hepatic Flexure | 7 (2.4) |
| Transverse Colon | 14 (4.9) |
| Splenic Flexure | 6 (2.1) |
| Descending Colon | 8 (2.8) |
| Sigmoid Colon | 71 (24.7) |
| Rectum | 133 (46.5) |
| Not Available | 2 (0.7) |
| Clinical Stage at presentation | |
| Tisa | 1 (0.3) |
| I | 28 (9.8) |
| II | 42 (14.7) |
| III | 98 (34.3) |
| IV | 107 (37.4) |
| Not Available | 10 (3.5) |
| Histology | |
| Well differentiated | 10 (3.5) |
| Moderately differentiated | 183 (64.0) |
| Poorly differentiated | 46 (16.1) |
| Signet cells | 2 (0.7) |
| Mucinous features | 13 (4.5) |
| Genetic Analysis | |
| Germline MLH 1 mutation | 8 (3.5) |
| Germline MSH 2 mutation | 5 (2.2) |
| Germline MSH 6 mutation | 3 (1.3) |
| Tumor MSIb High | 4 (1.7) |
| Tumor MLH1/PMS2 | 3 (1.3) |
| Tumor MSH2/MSH6 | 1 (0.4) |
| BRAF mutation | 3 (3.5) |
| KRAS mutation | 56 (40.3) |
Tis = in situ.
MSI = microsatellite instability.
Clinical stage at presentation data were available for all but ten patients. One patient with history of Crohn’s disease and rectal stricture with biopsy pathology demonstrating Tis proceeded to surgical resection where final pathology demonstrated stage III disease. Early stage at presentation included 28 stage I patients (10.1%) and 42 stage II patients (15.2%). The majority of patients presented at an advanced clinical stage, with 98 stage III patients (35.5%) and 107 stage IV patients (38.7%).
Tumor genetics were available for 255 specimens, with mismatch repair data available for 230. Microsatellite instability was present in 10.4% (n=24) of specimens. Pathogenic variant data demonstrated MLH1 deficiency in 8, MSH2 deficiency in 5, MSH6 deficiency in 3, and incomplete variant data for the remaining 8. BRAF testing was available for 85 specimens with mutated BRAF in three tumors. Five of the MLH1/PMS2-deficient specimens were tested for BRAF mutations, all were negative. Promotor hypermethylation data were unavailable for these specimens. Kirsten rat antigen sarcoma (KRAS) testing results were available in 139 patients with 40.3% tumors demonstrating mutated KRAS.
Symptomatology
Nearly all patients (95.1%) were symptomatic at diagnosis (Table 3). The most common presenting symptoms were changes in bowel habits and rectal bleeding (53.5 and 53.1%, respectively) followed by abdominal pain (46.9%); 84.6% of patients experienced at least one of these three symptoms. Constitutional symptoms including weight loss, fever, fatigue/malaise, and anorexia/early satiety were present in 44% of patients. Other abdominal symptoms included diarrhea (29.0%), constipation (19.4%), distention (11.5%), and upper gastrointestinal symptoms (26.6%). Less common symptoms included anemia (17.1%), back pain (7.3%), rectal pain (7.3%), obstruction (5.6%), urinary symptoms (4.5%), and pelvic pressure (2.1%). Twenty-seven patients were noted to attribute their symptoms to “hemorrhoids” (9.4%).
Table 3:
Symptomatology
| n (%) | |
|---|---|
| Symptom at Presentation | |
| Pain | 180 (62.9) |
| Back pain | 21 (7.3) |
| Rectal pain | 21 (7.3) |
| Abdominal pain | 134 (46.9) |
| Changes in bowel habits | 153 (53.5) |
| Rectal bleeding | 152 (53.1) |
| Weight loss | 91 (31.8) |
| Diarrhea | 83 (29.0) |
| Upper gastrointestinal symptoms | 76 (26.6) |
| Constipation | 56 (19.4) |
| Fatigue/malaise | 45 (15.7) |
| Bloating/distention | 33 (11.5) |
| Anorexia/early satiety | 34 (11.9) |
| “Hemorrhoids” | 27 (9.4) |
| Obstruction | 16 (5.6) |
| Fever | 14 (4.9) |
| Urinary symptoms | 13 (4.5) |
| Number of symptoms | |
| 0 symptoms | 14 (4.9) |
| 1 symptom | 40 (13.9) |
| ≥2 symptoms | 232 (81.1) |
| Duration of symptoms | |
| <1 months | 49 (18.9) |
| 1 – 3 months | 79 (30.5) |
| ≥3 months | 131 (50.6) |
| Not available | 27 (9.4) |
Eight patients were asymptomatic at the time of presentation with their cancers identified during screening colonoscopy or incidentally. Six were diagnosed via screening endoscopy performed for family history of CRC, or surveillance for personal history of another malignancy or inflammatory bowel disease. One patient was diagnosed at the time of elective bariatric procedure and one was diagnosed during a preoperative workup for an unrelated condition. Six additional patients did not have symptom data extractable from the medical record and were considered asymptomatic.
The number of symptoms at presentation ranged from 0 to 8. Patients typically had more than one symptom with 85.3% of patients presenting with two or more symptoms and 66.9% with three or more. Of the 259 patients for whom symptom chronicity was available, symptom duration prior to presentation ranged from days to years. One hundred and thirty-one patients (50.6%) had symptoms for three months or longer, 173 (66.8%) for two months or longer, 210 patients (81.1%) for one month or longer.
Symptomatology by Stage
Symptomatology amongst early versus later clinical stages did not demonstrate significant differences in either the number or duration of symptoms. Of patients presenting with stage I and II disease, 5.5% of patients were asymptomatic while 78.1% had two or more symptoms; in Stage III and IV, 4.4% of patients were asymptomatic while 83.5% with two or more symptoms. Duration of symptoms in stage I and II was greater than three or more months in 51.6% (n=32/62); in Stage III and IV, duration of symptoms was greater than three months in 51.1% (n=96/188, all p>0.01).
Symptomatology by Tumor Location
Symptoms differed depending on tumor location (Table 4). Right-sided tumors more often presented with abdominal pain, upper gastrointestinal symptoms, anemia, and anorexia, while left-sided tumors more often presented with rectal pain, rectal bleeding, and “hemorrhoid” complaints (all p<0.01). The majority of patients in each group presented with two or more symptoms. While the number of symptoms did not differ between left and right-sided cancer (p = 0.30), the duration of symptoms did (p<0.01). Left-sided cancers were associated with longer interval from symptom onset to diagnosis, with 56.8% of patients with left-sided cancers enduring symptoms for at least three months compared to 30.5% in right-sided cancers (p<0.01).
Table 4:
Symptomatology in Left versus Right-sided cancers
| Category | Left-sided (218), n (%) | Right-sided (66), n (%) | P value |
|---|---|---|---|
| Symptom | |||
| Any pain | 129 (59.2) | 50 (75.8) | 0.019 |
| Back pain | 16 (7.3) | 5 (7.6) | 1.0 |
| Rectal pain | 21 (9.6) | 0 (0) | <0.01 |
| Abdominal pain | 89 (40.8) | 44 (66.7) | <0.01 |
| Rectal bleeding | 135 (61.9) | 17 (25.8) | <0.01 |
| Changes in bowel habits | 122 (56.0) | 31 (47.0) | 0.21 |
| Weight loss | 69 (31.7) | 21 (31.8) | 1.0 |
| Diarrhea | 63 (28.9) | 20 (30.3) | 0.88 |
| Upper gastrointestinal symptoms | 46 (21.1) | 28 (42.4) | <0.01 |
| Constipation | 45 (20.6) | 10 (15.2) | 0.38 |
| Fatigue/malaise | 33 (15.1) | 12 (18.2) | 0.57 |
| Anemia | 29 (13.3) | 20 (30.3) | <0.01 |
| Bloating/distention | 28 (12.8) | 5 (7.6) | 0.28 |
| Anorexia/early satiety | 19 (8.7) | 15 (22.7) | <0.01 |
| “Hemorrhoids” | 26 (11.9) | 1 (1.5) | <0.01 |
| Obstruction | 8 (3.7) | 7 (10.6) | 0.05 |
| Fever | 10 (4.6) | 6 (9.1) | 0.22 |
| Urinary symptoms | 7 (3.2) | 5 (7.6) | 0.16 |
| Number of symptoms | p = 0.30 | ||
| 0 | 12 (5.5) | 2 (3.0) | |
| 1 | 28 (12.8) | 11 (16.7) | |
| 2 | 39 (17.9) | 11 (16.7) | |
| 3 | 54 (24.8) | 14 (21.2) | |
| 4 | 42 (19.3) | 10 (15.2) | |
| 5+ | 43 (19.7) | 18 (27.3) | |
| Duration of symptoms | p < 0.01 | ||
| < 1 month | 30 (15.1) | 18 (30.5) | |
| 1 – 3 months | 56 (28.1) | 23 (39.0) | |
| ≥ 3 months | 113 (56.8) | 18 (30.5) |
Symptoms in patients younger than 45 compared to 45 years of age and older
As recent guidelines have lowered the age to start CRC screening to 45 years, analysis was performed to identify differences in symptoms between patients younger than 45 and those 45 years or age and older. The quality, quantity, and duration of symptoms in younger cohort and older cohort did not differ statistically (all p>0.01).
Survival
At median follow-up of 55.1 months, median recurrence/progression-free (RPFS) and overall survival (OS) were 35.3 and 89.4 months, respectively. Overall survival at 5 years was 60.9% (95% confidence interval (CI) 54.8–67.8%), while survival without recurrence or progression at 5 years was 40.8% (CI 34.9–47.7%) (Figure 1).
Figure 1.
Survival in Young Colorectal Cancer Patients
To better understand correlations with survival outcomes, symptoms showing univariable association with RPFS and OS were entered into multivariable models and reverse stepwise selection removed those showing non-significant correlation until only those with independent association with RPFS or OS remained (Table 5). Factors associated with worse OS on univariable analysis included appetite loss and back pain (median 22.9 vs. 101.9 and 28.9 vs. 101.9 months), while patients with “hemorrhoids” had improved survival (5-year OS 95.6 vs. 57.4%). On multivariable analysis, back pain, urinary symptoms, and appetite loss were associated with worse recurrence/progression-free survival (p<0.05 for all). Patients with “hemorrhoids” had numerically better RPFS (5-year RPFS 67.2 vs. 37.9%), but this did not reach significance in the multivariable model (HR 0.51, p=0.05). Only appetite loss and back pain (HR 2.24 and 2.21) were independently associated with worse OS, although “hemorrhoids” and symptom duration >3 months showed trends towards improved survival (HR 0.33 and 0.55, p=0.06).
Table 5:
Factors associated with Disease Progression or Recurrence and Overall Survival on Multivariable Analysis
| Univariable | Multivariable | |||||||
|---|---|---|---|---|---|---|---|---|
| RPFS | OS | RPFS | OS | |||||
| Factor | HRa (CI)b | p-value | HR (CI) | p-value | HR (CI) | p-value | HR (CI) | p-value |
| Abdominal pain | 1.38 (1.01 – 1.90) | 0.04 | 1.39 (0.95 – 2.05) | 0.09 | - | - | - | - |
| Back Pain | 2.09 (1.26 – 3.46) | <0.01 | 2.46 (1.37 – 4.40) | <0.01 | 2.11 (1.27–3.51) | <0.01 | 2.21 (1.20–4.09) | 0.01 |
| Rectal bleeding | 0.72 (0.52 – 0.98) | 0.04 | 0.64 (0.44 – 0.95) | 0.02 | - | - | - | - |
| “Hemorrhoids” | 0.46 (0.24 – 0.91) | 0.03 | 0.23 (0.07 – 0.72) | <0.01 | 0.51 (0.26–1.00) | 0.05 | 0.33 (0.10–1.06) | 0.06 |
| Constitutional Symptoms | 1.46 (1.06 – 1.99) | 0.02 | 1.50 (1.02 – 2.20) | 0.04 | - | - | - | - |
| Urinary Symptoms | 2.24 (1.18 – 4.27) | 0.01 | 1.55 (0.68–3.53) | 0.3 | 2.25 (1.18–4.31) | 0.01 | - | - |
| >3 months | 0.80 (0.48–1.32) | 0.4 | 0.53 (0.29–0.98) | 0.04 | - | - | 0.55 (0.30–1.02) | 0.06 |
| Appetite Loss | 1.95 (1.27–3.00) | <0.01 | 2.61 (1.61–4.22) | <0.01 | 1.78 (1.15–2.75) | <0.01 | 2.24 (1.33–3.76) | <0.01 |
| Bowel Habit Changes | 0.75 (0.50–1.12) | 0.2 | 0.54 (0.31–0.96) | 0.03 | - | - | - | - |
| Fever | 1.97 (1.11–3.48) | 0.02 | 1.86 (0.94–3.69) | 0.07 | - | - | - | - |
| Weight Loss | 1.53 (1.11–2.11) | <0.01 | 1.46 (0.98–2.17) | 0.06 | - | - | - | - |
Right-sided tumors had worse RPFS (HR 1.56, CI 1.09–2.23, p=0.01) and OS (HR 1.89, CI 1.24–2.87, p<0.01) compared to left-sided cancers. Right-sided tumors were much more likely to be stage 4 upon diagnosis (54 vs. 34% of all patients, p<0.01), and rectal bleeding or “hemorrhoids” were more common in left-sided tumors (“hemorrhoids” in 12% of left-sided vs. 1.5% of right-sided tumors), suggesting that early recognition of bleeding from left-sided cancers may lead to diagnosis at an earlier stage, with resulting better survival.
Discussion
The rapid increase in CRC among young patients demands efforts to better identify affected individuals. While some guidelines have lowered the recommended age for CRC screening to 45 years,(25, 26) many patients develop disease earlier and would not be identified by screening alone. In our cohort of 286 patients, 55.6% of patients were younger than 45 years. Symptoms most frequently reported were rectal bleeding, change in bowel habits, and abdominal pain. Over 85.3% of patients had two or more symptoms, many of which lasted for months prior to diagnosis, and patients with appetite loss and back pain had worse overall survival on multivariable analysis. Identifying symptom patterns associated with CRC in young patients may help facilitate timely diagnosis.
Among studies evaluating the symptomatology of CRC in the young, the largest is a 2004 literature review by O’Connell et al. This reviewed 55 studies with 5,051 CRC patients diagnosed under the age of 40 and demonstrated, similar to our findings, that abdominal pain and rectal bleeding were the most common symptoms at presentation followed by weight loss and change in bowel habits.(8) More recently, Law et al. evaluated a cohort of 154 patients, and Riaz et al. evaluated a cohort of 105 patients. Both investigators reported abdominal pain and rectal bleeding as the most common symptoms.(4, 27) Rectal bleeding is not unique to young patients, as evidenced by a systematic review from Canada by Del Giudice et al. They found that in patients of all ages, the first episode of rectal bleeding had a positive predictive value (PPV) of 5.0% for presence of CRC, but when appearing alongside weight loss or change in bowel habits, its PPV was increased to 13.0 and 10.5%, respectively. Furthermore, rectal bleeding without perianal symptoms had a PPV of 10.8%.(28) This suggests that at least one in ten patients who present with rectal bleeding and associated symptoms or rectal bleeding without obvious perianal disease actually have cancer. Prospective data of relevance of rectal bleeding is limited; there is a small study including 290 patients of all ages that demonstrated a high incidence of pathologic findings on endoscopy, which increased with age, though overall detection of cancer was rare(29).
When evaluating symptom duration, we found most patients in our population had symptoms lasting at least three months. Another study found the duration of symptoms in young CRC patients ranged from two weeks to two years, with median duration of 6–12 months.(30, 31) This includes a small study with 53 young patients having symptom duration greater than three months in over 50% of patients, similar to our cohort. Unfortunately, there is limited literature on symptom duration in benign disease against which to contrast these results.
Often primary care providers (PCPs) first confront patients with these symptoms. One challenge in diagnosis is that symptoms associated with colon cancer outcomes in the young, such as “hemorrhoids” and back pain, are common in this age group and usually result from benign processes. To help with this conundrum, guidelines exist to advise PCPs on when to pursue further testing. Canadian expert committee guidelines note that rectal bleeding independently demonstrated a median PPV for CRC of 9.7% and changes in bowel habits a median PPV of 7.5%. When these symptoms were combined the median PPV increased to 10.5%.(32) The authors concluded that patients with concerning symptoms (rectal bleeding, change in bowel habits, weight loss, abdominal pain, perianal symptoms, or anemia) should be considered for referral to a specialist if not resolved in 4–6 weeks. Similarly, the American Academy of Family Physicians (AAFP) 2019 guidelines include considering colonoscopy in patients with unexplained gastrointestinal bleeding, iron deficiency anemia, or unexplained significant diarrhea.(26, 33) Our study suggests adherence to these guidelines would capture many of the young patients affected, particularly those with left-sided cancers who most commonly present with rectal bleeding and changed bowel habits. Our results further highlight attention to appetite loss, which is both less common in young patients and associated with survival outcomes.
The literature demonstrates a predominance of left-sided cancers in young patients.(2, 4, 6, 27, 34) This was similar in our cohort, with 76.8% of patients having left-sided neoplasms. Evaluation of symptoms based on anatomic locations in our study population revealed right-sided neoplasms most often had abdominal pain and left-sided had rectal bleeding. This phenomenon was also seen by Riaz et al. with a cohort of 105 young patients with CRC.(27) Right-sided cancers more often had anemia, suggesting that tumors on both sides bleed, but that bleeding may be recognized more often when coming from left-sided cancers.
Survival analysis demonstrated that appetite loss, back pain, and constitutional symptoms were associated with worse overall survival, shorter time to progression and increased risk of recurrence whereas rectal bleeding and longer duration of symptoms were associated with better outcomes. Whether presenting symptoms relate to prognosis remains unclear. In a study by Minjoung et al. looking at solid tumors in a 2014 English National Cancer Diagnosis Audit, they found neck lump, chest pain, and back pain to be associated with an increased odds of patients having stage IV cancer, but not for other symptoms examined including rectal bleeding, urinary symptoms, change in bowel habits, or abdominal pain.(35) A study by Jullumstrø et al. analyzing symptoms and survival among 4,155 patients diagnosed with colon or rectal cancer in Norway from 1980 to 2004 found that, similar to our results, prolonged duration was correlated to improved survival.(36) They also found an association between increased TNM stage and prolonged duration, which was not the case in our study. This may be due to the population in their study, which included patients of all ages.
Interestingly, we found that patients with longer duration of symptoms before diagnosis showed a trend towards improved overall survival, consistent with previous data. Jullumstrø et al. evaluated 4,155 patients with colon and rectal cancer and found patients with increased duration of symptoms had improved 4 and 5-year survival, similar to our cohort(36). While this seems counterintuitive, a long time to diagnosis may reflect indolent biology and immortal-time bias in this subset of patients. In other words, aggressive tumors produce severe symptoms or death quickly, which prevents patients from having a long time to diagnosis. Only slow-growing tumors can produce symptoms for many months, and these tumors have longer survival times.
Due to the nature of our study, we did not perform comparison to an older subset of patients. We did find that symptoms were similar when comparing patients younger than 45 to those aged 45–50 in our cohort. Extrapolation from existing data in the literature mostly includes all ages, though with high median ages. The majority come from studies comparing right sided and left sided cancers. Right sided cancers are cited to more commonly present with anemia, while left sided cancer present with rectal bleeding and changes in bowel habits. Additional common symptoms include abdominal pain and bloating (37–40).
Our study provides rich data to characterize symptoms of CRC in young patients. Strengths of this study include a large patient population from a tertiary center with widespread referral base. Specific datapoints were able to be assessed in this institutional study that would otherwise not be available in a national database, including patient symptoms, duration, and time to progression of disease or recurrence. Limitations of this study include its retrospective nature and using a single institution’s electronic medical record. Reported symptoms and duration often relied on patient recall which is subject to inaccuracy.
Conclusion
In this single institutional study, symptoms most prominently associated with CRC in young patients were rectal bleeding, change in bowel habits, and abdominal pain. The majority of patients had two or more symptoms enduring at least three months prior to their cancer diagnosis. Based on our results, patients presenting with persistent and multiple symptoms need to be considered for formal CRC testing. With increased awareness and vigilant attention to symptomatology, we may be able to diagnose young patients with CRC in a more timely fashion.
Funding:
This work was supported by NIH T32#CA078586 (SKS).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest: The authors have no relevant financial disclosures.
This work was presented at the American Society of Colorectal Surgeons Annual Virtual Scientific Meeting, June 6–10, 2020.
Availability of Data and Material:
Available under reasonable request.
References
- 1.Siegel RL, Fedewa SA, Anderson WF, et al. Colorectal Cancer Incidence Patterns in the United States, 1974–2013. JNCI: Journal of the National Cancer Institute. 2017;109(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Liang J, Kalady MF, Church J. Young age of onset colorectal cancers. Int J Colorectal Dis. 2015. Dec;30(12):1653–7. Epub 2015/09/12. eng. [DOI] [PubMed] [Google Scholar]
- 3.Colorectal Cancer Facts & Figures 2017–2019. Atlanta: American Cancer Society; [updated 2017]; Available from: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/colorectal-cancer-facts-and-figures/colorectal-cancer-facts-and-figures-2017-2019.pdf.
- 4.Law JH, Koh FH, Tan KK. Young colorectal cancer patients often present too late. Int J Colorectal Dis. 2017. Aug;32(8):1165–9. Epub 2017/05/20. eng. [DOI] [PubMed] [Google Scholar]
- 5.Jain M, FACG Ankur and Jain MD, FACG Shilpa. Rising Incidence of Colorectal Cancer in Patients Younger than Age 50 in Hawai’i. Hawaii J Med Public Health. 2019;78(6):195–9. [PMC free article] [PubMed] [Google Scholar]
- 6.Mauri G, Sartore-Bianchi A, Russo AG, et al. Early-onset colorectal cancer in young individuals. Mol Oncol. 2019. Feb;13(2):109–31. Epub 2018/12/07. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.O’Connell JB, Maggard MA, Liu JH, et al. Rates of colon and rectal cancers are increasing in young adults. Am Surg. 2003. Oct;69(10):866–72. Epub 2003/10/23. eng. [PubMed] [Google Scholar]
- 8.O’Connell JB, Maggard MA, Livingston EH, Yo CK. Colorectal cancer in the young. Am J Surg. 2004. 2004/03//;187(3):343–8. [DOI] [PubMed] [Google Scholar]
- 9.Siegel RL, Jemal A, Ward EM. Increase in incidence of colorectal cancer among young men and women in the United States. Cancer Epidemiol Biomarkers Prev. 2009. Jun;18(6):1695–8. Epub 2009/06/10. eng. [DOI] [PubMed] [Google Scholar]
- 10.Ahnen DJ, Wade SW, Jones WF, et al. The increasing incidence of young-onset colorectal cancer: a call to action. Mayo Clin Proc. 2014. Feb;89(2):216–24. Epub 2014/01/08. eng. [DOI] [PubMed] [Google Scholar]
- 11.Lui RN, Tsoi KKF, Ho JMW, et al. Global Increasing Incidence of Young-Onset Colorectal Cancer Across 5 Continents: A Joinpoint Regression Analysis of 1,922,167 Cases. Cancer Epidemiol Biomarkers Prev. 2019. Aug;28(8):1275–82. Epub 2019/05/23. eng. [DOI] [PubMed] [Google Scholar]
- 12.Siegel RL, Jakubowski CD, Fedewa SA, et al. Colorectal Cancer in the Young: Epidemiology, Prevention, Management. Am Soc Clin Oncol Educ Book. 2020. Mar;40:1–14. Epub 2020/04/22. eng. [DOI] [PubMed] [Google Scholar]
- 13.Abdelsattar ZM, Wong SL, Regenbogen SE, et al. Colorectal cancer outcomes and treatment patterns in patients too young for average-risk screening. Cancer. 2016. Mar 15;122(6):929–34. Epub 2016/01/26. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kearney DE, Cauley CE, Aiello A, et al. Increasing Incidence of Left-Sided Colorectal Cancer in the Young: Age Is Not the Only Factor. J Gastrointest Surg. 2020. 2020/10/01;24(10):2416–22. [DOI] [PubMed] [Google Scholar]
- 15.Segev L, Kalady MF, Church JM. Left-Sided Dominance of Early-Onset Colorectal Cancers: A Rationale for Screening Flexible Sigmoidoscopy in the Young. Dis Colon Rectum. 2018. Aug;61(8):897–902. Epub 2018/05/18. eng. [DOI] [PubMed] [Google Scholar]
- 16.Ballester V, Rashtak S, Boardman L. Clinical and molecular features of young-onset colorectal cancer. World J Gastroenterol. 2016;22(5):1736–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Siegel RL, Miller KD, Fedewa SA, et al. Colorectal cancer statistics, 2017. CA Cancer J Clin. 2017. May 6;67(3):177–93. Epub 2017/03/02. eng. [DOI] [PubMed] [Google Scholar]
- 18.You YN, Lee LD, Deschner BW, Shibata D. Colorectal Cancer in the Adolescent and Young Adult Population. JCO Oncology Practice. 2020;16(1):19–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Siegel RL, Miller KD, Jemal A. Colorectal Cancer Mortality Rates in Adults Aged 20 to 54 Years in the United States, 1970–2014. JAMA. 2017. Aug 8;318(6):572–4. Form for Disclosure of Potential Conflicts of Interest and none were reported. Epub 2017/08/09. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Bleyer A CAUTION! Consider cancer: common symptoms and signs for early detection of cancer in young adults. Semin Oncol. 2009. Jun;36(3):207–12. Epub 2009/05/23. eng. [DOI] [PubMed] [Google Scholar]
- 21.Dozois EJ, Boardman LA, Suwanthanma W, et al. Young-onset colorectal cancer in patients with no known genetic predisposition: can we increase early recognition and improve outcome? Medicine (Baltimore). 2008. Sep;87(5):259–63. Epub 2008/09/17. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B (Methodological). 1995;57(1):289–300. [Google Scholar]
- 23.Kaplan EL, Meier P. Nonparametric Estimation from Incomplete Observations. Journal of the American Statistical Association. 1958;53(282):457–81. [Google Scholar]
- 24.Schemper M, Smith TL. A note on quantifying follow-up in studies of failure time. Control Clin Trials. 1996. Aug;17(4):343–6. Epub 1996/08/01. eng. [DOI] [PubMed] [Google Scholar]
- 25.American Cancer Society Guideline for Colorectal Cancer Screening. American Cancer Society; 2020. [updated November 17, 2020; cited 2021 May 12].
- 26.Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA Cancer J Clin. 2018;68(4):250–81. [DOI] [PubMed] [Google Scholar]
- 27.Riaz R, Masood N, Benish A. Red flag symptoms: detailed account of clinicopathological features in young-onset colorectal cancer. Intest Res. 2017;15(2):203–7. Epub 2017/04/27. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Del Giudice ME, Vella ET, Hey A, et al. Systematic review of clinical features of suspected colorectal cancer in primary care. Can Fam Physician. 2014. Aug;60(8):e405–15. Epub 2014/08/15. eng. [PMC free article] [PubMed] [Google Scholar]
- 29.Koning MV, Loffeld R. Rectal bleeding in patients with haemorrhoids. Coincidental findings in colon and rectum. Fam Pract. 2010;27(3):260–2. [DOI] [PubMed] [Google Scholar]
- 30.Chan KK, Dassanayake B, Deen R, et al. Young patients with colorectal cancer have poor survival in the first twenty months after operation and predictable survival in the medium and long-term: Analysis of survival and prognostic markers. World J Surg Oncol. 2010. 2010/09/15;8(1):82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Campos FG. Colorectal cancer in young adults: A difficult challenge. World J Gastroenterol. 2017;23(28):5041–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Del Giudice ME, Vella ET, Hey A, et al. Guideline for referral of patients with suspected colorectal cancer by family physicians and other primary care providers. Can Fam Physician. 2014. Aug;60(8):717–23, e383–90. Epub 2014/08/15. eng fre. [PMC free article] [PubMed] [Google Scholar]
- 33.Colonoscopy (Position Paper). American Academy of Family Physicians.
- 34.Kasi PM, Shahjehan F, Cochuyt JJ, et al. Rising Proportion of Young Individuals With Rectal and Colon Cancer. Clin Colorectal Cancer. 2019. 2019/03/01/;18(1):e87–e95. [DOI] [PubMed] [Google Scholar]
- 35.Koo MM, Swann R, McPhail S, et al. Presenting symptoms of cancer and stage at diagnosis: evidence from a cross-sectional, population-based study. The Lancet Oncology. 2020;21(1):73–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Jullumstrø E, Lydersen S, Møller B, et al. Duration of symptoms, stage at diagnosis and relative survival in colon and rectal cancer. Eur J Cancer. 2009. Sep;45(13):2383–90. Epub 2009/04/10. eng. [DOI] [PubMed] [Google Scholar]
- 37.Holtedahl K, Borgquist L, Donker GA, et al. Symptoms and signs of colorectal cancer, with differences between proximal and distal colon cancer: a prospective cohort study of diagnostic accuracy in primary care. BMC Fam Pract. 2021. Jul 8;22(1):148. Epub 2021/07/10. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Hamilton W Cancer diagnosis in primary care. Br J Gen Pract. 2010. Feb;60(571):121–8. Epub 2010/02/06. eng. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kanellos D, Kitsios G, Kanellos I, et al. Anaemia as a symptom of right colon cancer. Tech Coloproctol. 2004. Nov;8 Suppl 1:s62–4. Epub 2005/01/19. eng. [DOI] [PubMed] [Google Scholar]
- 40.Alexiusdottir KK, Möller PH, Snaebjornsson P, et al. Association of symptoms of colon cancer patients with tumor location and TNM tumor stage. Scand J Gastroenterol. 2012. Jul;47(7):795–801. Epub 2012/04/18. eng. [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
Available under reasonable request.