DOI: https://doi.org/10.21203/rs.3.rs-2413981/v1
Objective: Early-onset colorectal cancer (EOCRC) has different clinical and pathological characteristics compared to late-onset CRC (LOCRC). Mortality rate as a post-operative outcome is a patient's postoperative outcome considered based on the state of life or death. The objective of this research is to analyze the comparison between clinicopathological aspect of early-onset vs late-onset CRC as well as their correlation with the mortality rate in Indonesia to support global data. We performed a case-control study on 170 subjects with CRC from November 2021- November 2022 in Dr. Hasan Sadikin General Hospital. Data were extracted from electronic medical records Colorectal Cancer (CRC) Registry. Bivariate and correlation analyses were used to analyse the difference between variables using IBM SPSS 24.0. P<0.05 was considered statistically significant.
Result: Anemia and tumor location variables were significantly different in the early-onset group compared to the late-onset group (P<0.001). It was also found that anemia (P<0.001), pathological features (P<0.001), and tumor location (P=0.013) had significantly low correlation with onset of CRC (r=0.325; r=0.397; r=0.342, respectively). There is no statistically significant correlation between the clinicopathological features of CRC in both onset and mortality rates in this study.
Based on the patient's age at the first time diagnosed, colorectal cancer (CRC) is divided into two groups, which are early-onset colorectal cancer (EOCRC) and late-onset colorectal cancer (LOCRC).1 Amsterdam criteria stated that EOCRC refers to subjects diagnosed at < 50 years of age while LOCRC refers to subjects diagnosed at the age of ≥ 50. While most common cases of CRC represent in older subjects, the incidence of EOCRC tends to rise globally.1,2
EOCRC has different clinical characteristics compared to LOCRC. The majority of early-onset subjects have red flags such as unexplained anemia, rectal bleeding, and changes in bowel habits.3 LOCRC has a more diverse range of symptoms including rectal bleeding/haematochezia, melena, constipation, nausea, diarrhea, abdominal pain, presence of an abdominal mass, anemia, unexplained fatigue, and weight loss. In LOCRC, the subjects come to see the physician earlier since the symptoms felt to be more disturbing. Thus, LOCRC tends to be diagnosed at an early stage.3–5 In many cases, EOCRC is diagnosed with metachronous cancer, showing more advanced stages such as stage III and stage IV. EOCRC is more likely to cause metastases during the course of the disease compared to LOCRC.6–8
EOCRC has different histopathological characteristics compared to LOCRC. EOCRC tends not to have precursor adenomatous lesions and more signet ring cells are found, whereas in late-onset there is no signet ring cell at all. A study in Indonesia showed that the histopathological features of CRC were different from previous studies abroad. In Dr. Hasan Sadikin General Hospital Bandung, the most frequent histopathological features of subjects aged less than 40 years are well-differentiated, moderately differentiated, poorly differentiated adenocarcinoma, and signet ring cells.9 The majority of the histopathological features were similar to those of LOCRC, namely adenocarcinoma (89.8%).10
For both onsets, surgery is the main modality for early stage cancer with curative goals. In surgical therapy, many factors determine the success rate such as clinical conditions, histopathological features, patient comorbidities, and post-operative complications. Comorbidities in subjects can increase the risk of postoperative mortality rate.1,11 Mortality rate as a postoperative outcome is a patient's postoperative outcome based on the condition of life or death.12 Nowadays, there have only been studies abroad that have compared the clinicopathological characteristics of early-onset and late-onset colorectal cancer (EOCRC vs LOCRC)13–15, but there is a lack of data that analyse the correlation between these clinicopathological aspects and post-operative outcomes such as mortality rates, especially in developing countries like Indonesia. Therefore, this study aimed to analyse the comparison of clinicopathological features between early-onset and late-onset CRC as well as the correlation between CRC onset and mortality rate in Dr. Hasan Sadikin Hospital, Bandung as local data that can be used to support global data.
This was a cross-sectional study designed to examine the comparison between the clinicopathological aspect of early-onset vs late-onset CRC as well as their correlation with the mortality rate in CRC subjects as the primary outcomes. The data used were secondary data extracted from the medical records of subjects at CRC Registry, Dr. Hasan Sadikin General Hospital Bandung, starting from November 2021- November 2022 until the minimum number of samples was fulfilled using a consecutive sampling method.
(1) CRC subjects who had been diagnosed clinically, radiologically and, pathologically at Dr. Hasan Sadikin General Hospital Bandung and (2) the subjects that had definitive therapies only at Dr. Hasan Sadikin General Hospital Bandung were included in this study. The subjects with incomplete medical record data were excluded from this study.
The data collection from CRC Registry was categorized as low-risk as it was conducted using electronic medical record data. After receiving approval and recommendations from the Ethics Committee Review Board of Hasan Sadikin General Hospital - Faculty of Medicine, Universitas Padjadjaran, all procedures were performed in accordance with applicable guidelines and regulations, with reference number LB.02.01/X.6.5/328/2022.
Comparison analysis would be carried out using the independent T-test, Chi-Square, Fisher's Exact, Mann-Whitney, and Kolmogorov-Smirnov test depending on the distribution of data. Correlation analysis would be utilized as the analysis between variables. A P-value of < 0.05 was considered statistically significant.
Subject Characteristics
As shown in Table 1, this study involved 170 subjects aged 20–89 years with a mean age of 52.35 ± 14.31 years. In this study, there were more female subjects (57.06%) than male subjects (42.94%). Of all subjects, the majority of subjects (134 subjects) complained of difficult bowel movements or constipation (78.82%). In addition, in general, subjects also experienced hematochezia (48.82%), anemia based on clinical symptoms and laboratory results (34.71%), abdominal pain (23.53%), and diarrhea in 14.12% patients.
Table 1. Subject Characteristics
|
Variables |
Total Subjects (n=170) |
Percentage (%) |
|
Age (Years) Mean ± SD Median (Min – Max) |
52.35 ± 14.31 51.5 (20 – 89) |
|
|
Sex Male Female |
73 97 |
42.94 57.06 |
|
Clinical Features Hematochezia Melena Abdominal Pain Constipation Abdominal Mass Diarrhea Anemia |
83 1 40 134 3 24 59 |
48.82 0.59 23.53 78.82 1.76 14.12 34.71 |
|
Pathological Features Adenocarcinoma Mucinous Adenocarcinoma Signet Ring Cells Others |
124 4 20 22 |
72.94 2.35 11.76 12.94 |
|
Stages Stage 0 Stage I Stage IIA Stage IIB Stage IIC Stage IIIA Stage IIIB Stage IIIC Stage IV |
1 17 6 14 0 32 36 3 61 |
0.59 10.00 3.53 8.24 0.00 18.82 21.18 1.76 35.88 |
|
Tumor Locations Iliocecal Caecum Ascenden Colon Transverse Colon Descenden Colon Sigmoid Colon Rectosigmoid Rectum Anorectal Colon Cervical Combination |
1 0 21 10 14 23 17 75 1 1 1 6 |
0.59 0.00 12.35 5.88 8.24 13.53 10.00 44.12 0.59 0.59 0.59 3.53 |
|
Outcome Life Death |
159 11 |
93.53 6.47 |
|
Onset Early Late |
72 98 |
42.35 57.65 |
Note: In descriptive data, nominal-categoric data (age) is presented in the form of mean ± SD and median (min-max) while the others are presented in the form of percentages.
A comparative analysis (bivariate analysis) was performed as shown in Table 2. It was found that anemia and tumor location were significantly different in the early-onset group compared to the late-onset group (P < 0.001) while the other variables were not.
|
Variabel |
Early Onset |
Late Onset |
P-Value |
||
|
Subjects (n=72) |
Percentage (%) |
Subjects (n=98) |
Percentage (%) |
||
|
Sex Male Female |
36 36 |
50.00 50.00 |
37 61 |
37.76 62.24 |
0.111 |
|
Clinical Features Hematochezia Melena Abdominal Pain Constipation Abdominal Mass Diarrhea Anemia |
39 0 16 55 0 12 34 |
54.17 0.00 22.22 76.39 0.00 16.67 47.96 |
44 1 24 79 3 12 17 |
44.90 1.02 24.49 80.61 3.06 12.24 16.67 |
0.232 1.000 0.731 0.505 0.263 0.413 *<0.001 |
|
Pathological Features Adenocarcinoma Mucinous Adenocarcinoma Signet Ring Cells Others |
44 2 20 6 |
61.11 2.78 27.78 8.33 |
80 2 0 16 |
81.63 2.04 0.00 16.33 |
0.061 |
|
Stages Stage 0 Stage I Stage IIA Stage IIB Stage IIC Stage IIIA Stage IIIB Stage IIIC Stage IV |
0 7 2 8 0 8 18 1 28 |
0.00 9.72 2.78 11.11 0.00 11.11 25.00 1.39 38.89 |
1 10 4 6 0 24 18 2 33 |
1.02 10.20 4.08 6.12 0.00 24.49 18.37 2.04 33.67 |
0.386 |
|
Tumor Locations Iliocecal Caecum Ascenden Colon Transvers Colon Descenden Colon Sigmoid Colon Rectosigmoid Rectum Anorectal Colon Cervical Combination |
0 0 13 6 11 12 5 22 0 0 0 3 |
0.00 0.00 18.06 8.33 15.28 16.67 6.94 30.56 0.00 0.00 0.00 4.17 |
1 0 8 4 3 11 12 53 1 1 0 4 |
1.02 0.00 8.16 4.08 3.06 11.22 12.24 54.08 1.02 1.02 0.00 4.08 |
*0.001 |
|
Outcome Life Death |
66 6 |
91.67 8.33 |
93 5 |
94.90 5.10 |
0.531 |
Notes: Comparative analysis between clinicopathology and onset was carried out using the Independent T-Test method for numerical variables, Fisher's Exact for nominal categories and Mann-Whitney for ordinal categories if the data were normally distributed. For data that was not normally distributed, analysis was carried out using the Kolmogorov-Smirnov. The data was stated to be significant if the P-value *<0.05
It was found that anemia (P < 0.001), histopathological features (P < 0.001), and tumor location (P = 0.013) had significantly low correlation (respectively r = 0.325; r = 0.397; respectively). r = 0.342 respectively) for early-onset and late-onset CRC as shown in Table 3. There is no statistically significant correlation between the clinicopathological features of CRC in both onset and mortality rates in this study.
|
Variable |
Correlation |
r |
P Value |
|---|---|---|---|
|
Sex |
Phi |
0.122 |
0.111 |
|
Hematochezia |
Phi |
-0.092 |
0.232 |
|
Melena |
Phi |
0.066 |
0.390 |
|
Abdominal Pain |
Phi |
0.026 |
0.731 |
|
Constipation |
Phi |
0.051 |
0.505 |
|
Abdominal Mass |
Phi |
0.115 |
0.134 |
|
Diarrhea |
Phi |
-0.063 |
0.413 |
|
Anemia |
Phi |
0.325 |
*<0.001 |
|
Pathological Features |
Contingency Coefficient |
0.397 |
*<0.001 |
|
Stages |
Theta |
0.075 |
0.386 |
|
Tumor Locations |
Contingency Coefficient |
0.342 |
*0.013 |
|
Mortality Rate (Outcome) |
Phi |
-0.065 |
0.397 |
| Notes: P-value was statistically significant if *< 0.05. r = correlation coefficient (r = 0.20–0.40: low correlation; r = 0,40 − 0.70: moderate correlation; r = 0.70–0.90: high correlation; 0.90-1.00: very high correlation) | |||
In this study, EOCRC had the most tumor locations in the rectum and proximal colon while LOCRC had the most tumor locations in the distal colon. This study demonstrated significantly more clinical anemia in EOCRC patients with predominant tumor locations in the proximal colon relative to the distal colon and rectum. These findings confirm the results of several previous studies.5,16 The fecal occult blood test (FOBT) screening has been reported to have high sensitivity for detecting tumors of the colon and rectum, indicating that both colon and rectal cancers frequently bleed into the lumen. The distinction between proximal and distal colorectal cancer may be mechanically related to bleeding, but other effects such as immunological mechanisms need to be considered as well. 5,16,17
Proximal colon tumors often have distinct genetic characteristics (in particular, the BRAF V600E mutation and MMR deficiency/mismatch repair deficiency), which result from the development of serrated precursor lesions via the serrated route of colorectal carcinogenesis. 16,18 Patients with MMR-deficient tumors have a very high prevalence of anemia (72.5%), and both microcytic and normocytic anemia were more common in the MMR-deficient subgroup. However, MMR-deficient tumors were predominantly located in the proximal colon. In addition, MMR deficiency was not a significant predictor of blood hemoglobin levels in multiple linear regression. Further studies are needed to reliably analyze blood haemoglobin levels in MMR deficiency cases at different tumor sites.
Blood haemoglobin levels in colorectal cancer are inversely related to systemic inflammation. High serum IL-8, and low serum albumin, are mainly associated with normocytic anemia. IL-8 is a proinflammatory chemokine associated with the promotion of neutrophil chemotaxis and degranulation. Serum IL-8 levels are elevated in many malignancies and IL-8 is thought to be an important contributor to cancer-associated inflammation. Serum albumin levels reflect systemic inflammation since albumin synthesis decreases in response to IL-6. These findings support the notion that, in particular, normocytic anemia in colorectal cancer is associated with systemic inflammation. These associations may also have therapeutic significance, as modulation of the inflammatory response has shown some potencies of inflammatory anemia treatment.5,16,18
The patient mortality rate increased in advanced stages (stage IV, IIIA, and IIIB) in both the early-onset and late-onset CRC groups. This is in line with previous studies. The 5-year survival rate ranges from 77% in stage I colorectal cancer cases to only 2% in stage IV CRC cases. In developed countries, the five-year life expectancy is over 60%, but it is lower to < 50% in Iran. However, the 5-year survival for all cases of CRC increased from 42.7–44.6%, although not statistically significant (p = 0.76).19 Meanwhile, according to data from other centers in Indonesia, out of 142 CRC patients included in the study, 43% of patients survived during the observation period (five years). By analysis, subjects aged < 45 years showed a better 5-year survival (47.4% vs 41.3%).20 However, the study found no significant difference between age and survival. This finding differs from a study by Chao-Hsien et al which showed age to be a predictor of survival and prognosis in colorectal cancer.20 Early stages show a better prognosis whereas, in those with advanced stages, the cancer develops very progressive and aggressively so that it can reduce survival rates. 5,16−20
Based on the correlation analysis between variables, it was found that anemia, histopathological features, and tumor location significantly had a low correlation with the incidence of CRC onset. Taken together, these findings support the notion that, in particular, normocytic anemia in CRC is associated with systemic inflammation. Anemia is more common in EOCRC with the predominant tumor location in the proximal tumor area. This condition is related to genetic and molecular factors as previously described.
Based on the histopathological aspect, in theory, as many as 90% of CRC cases originate from the glandular mucosa and belong to the type of adenocarcinoma originating from the colorectal mucosal epithelial cells.5,21 Adenocarcinoma is characterized by the formation of glands. EOCRC has different pathological features compared to LOCRC. EOCRC tends not to have precursor adenomatous lesions and signet ring cell is common, whereas in late-onset there is no signet ring cell at all.22,23 This is in line with what was found in this study.
Signet ring cells show an infiltrative pattern or there is a collection of extracellular mucin and is classified as poorly differentiated.22 In other studies, EOCRC showed a feature of gland formation around 50–95% and was moderately differentiated.9,10 Left colorectal cancer is more common compared to the right and is associated with a worse pathological features and prognosis. On the other hand, the majority of cases of EOCRC are diagnosed as more advanced stages. EOCRC shows a lower survival rate compared to LOCRC patients. In this present study, adenocarcinoma was the dominant pathological feature in both groups as has been reported in other studies. 5,21,22,23 Other studies have also reported a higher prevalence of poorly differentiated histological signet ring cells in young colorectal patients presenting with aggressive tumors. However, in this study, there is no statistically significant correlation between the clinicopathological features of CRC in both onset and mortality rates.
There was a statistically significant difference in clinicopathological features (anemia and tumor location) between subjects EOCRC vs LOCRC. There is no statistically significant correlation between the clinicopathological features of CRC in both onset and mortality rates in this study.
Nevertheless, this study also has several limitations, including cross-sectional method derived from electronic medical records, thus only documented details were available for evaluation. Many relevant data such as details about adenoma, family history, and type of surgery were not adequately available. This study is also still monocentric. Nonetheless, this study provides relevant data regarding the clinicopathological features of EOCRC vs LOCRC as well as their correlation with mortality rate and will add to existing world databases to create valid conclusions.
|
CRC |
: |
Colorectal Cancer |
|
EOCRC |
: |
Early-Onset Colorectal Cancer |
|
FOBT |
: |
Fecal Occult Blood Test |
|
LOCRC |
: |
Late-Onset Colorectal Cancer |
|
MMR |
: |
Mismatch Repair |
Ethics Approval and Informed Consent to Participate
Ethics Committee Review Board of Hasan Sadikin General Hospital – Faculty of Medicine, Universitas Padjadjaran waived the need for informed consent with reference number LB.02.01/X.6.5/328/2022. All methods were carried out in accordance with relevant guidelines and regulations.
Consent for publication
Not Applicable
Availability of data and materials section
The authors declare that the personal data from any subjects involved in this study will not be shared based on subjects’ confidentialities.
Competing Interest
The authors have declared that no competing interest exist.
Funding
No funding
Authors’ contributions
KL, AM, RR, and EP did the conception of the study and revised the manuscript critically for important intellectual content.
AM and ADN did the acquisition of data, analysis and interpretation of the data, drafted the manuscript and revised the manuscript critically for important intellectual content.
Acknowledgements
No applicable.