Translate this page into:
A Study of Molecular Subtypes (Profile) of Colorectal Cancer and Their Correlation with Clinical and Pathological Profile in a Tertiary Care Center in India
, Sumit Goyal1, Anurag Mehta2,3,4, Manish Sharma1,, Venkata Pradeep Babu Koyyala1, Pankaj Goyal1,, Chaturbhuj Aggarwal1, M. Shiv Shankara Swamy1, Amrith B Patel1, Shrinidhi Nathany5, Moushumi Suryavanshi6,7, Anurag Sharma4, Satya Narayan Saraswat1, Satyajeet Soni1, Arpit Jain1, Pallavi Redhu1, Vineet Talwar1, D C Doval1
-
Received: ,
Accepted: ,
This article was originally published by Thieme Medical and Scientific Publishers Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Abstract
Background
Colorectal cancer (CRC) is a heterogeneous disease morphologically, histologically, and molecularly. Most of the studies are on this molecular heterogeneity and their clinicopathological correlation from the western world. Very few studies have been done in India.
Objectives
The aims of this study were to evaluate the clinical and pathological profile of CRCs, to determine the frequency of molecular subtypes of CRCs, to correlate between the molecular subtypes and their clinicopathological features, and to determine the association between different molecular subtypes of CRC.
Materials and Methods
A prospective noninvasive interventional study was done on 50 patients (both outpatients and inpatients) with newly diagnosed CRCs presenting to the Rajiv Gandhi Cancer Institute and Research Centre, Rohini, Delhi, from February 2019 to March 2020. Clinical and histopathological data were collected from case sheets as per the study proforma: history and physical examination, noninvasive and invasive imaging, and histopathological reports. Patients in whom tissue was insufficient or not available for testing for at least three of five molecular markers (KRAS, NRAS, BRAF, MSI, and MLH1 methylation) were excluded. The results were analyzed with SSPS 23.0 software. For comparison of the frequencies among groups, the chi-squared test and the Fisher exact test were used. A p-value of less than 0.05 was considered statistically significant.
Results
The median age was 53 years. The majority of the males (54%) had CRC and 44% were right-sided colon tumors. Of the 50 patients with CRC, 40, 0, 4, and 4% had KRAS mutation, NRAS and BRAF mutation, and deficient mismatch repair (dMMR), respectively. KRAS mutation was significantly associated with upfront liver metastases (p = 0.02) and well/moderate differentiation (p = 0.02). BRAF wild-type tumors were likely to be well differentiated (p = 0.02), and moreover, half of them (52%) had MLH1 promoter methylation. The proportion of dMMR was higher in male patients (p = 0.04). Deficient mismatch repair was associated with well/moderate differentiation (p = 0.02), early stage (p =0.02), and mild peritumoral lymphocytes (p = 0.01). None of the dMMR patients had stage IV CRC. In all, 27% of the patients (3/11) with dMMR tumors had germline mutation of the dMMR genes. The majority of dMMR tumors (43%, 3 out of 7) had MLH1 promoter methylation. Overall, 45% (5/11) of dMMR tumors harbored KRAS mutation.
Conclusion
In conclusion, this is a prospective study evaluating the correlations between RAS/BRAF mutation and dMMR status with clinicopathological characteristics in Indian CRC patients, which is slightly similar to worldwide reports with some exceptions. To the best of our knowledge, this is the first study to evaluate the molecular marker combinations in CRC in India.
Keywords
Introduction
The incidence of colorectal carcinoma (CRC) worldwide is 19.7 per 1,00,0000 population, with 23.6/1,00,000 in males and 16.3/1,00,000 in females.1 According to GLOBOCAN 2018,1 CRC ranks third in worldwide newly diagnosed cancer cases and accounts for 9.2% cancer related deaths. It is also the third most common cause of cancer-specific mortality in the Asian continent.1
CRC is a biologically, histologically, and epidemiologically heterogeneous disease. Recent studies showed that the molecular profile of CRC is also different according to the tumor site.23 Many studies were done in the western world and few Asian countries on the different molecular types of CRCs and their correlation clinically and pathologically. Somatic MLH1 DNA methylation is associated with older age, females, proximal tumors, and more likely to be BRAF V600E mutated.4 The pathological features with MSI-H CRC are mucinous histology (predominantly signet ring cell type), plenty of tumor-infiltrating lymphocytes, and poor differentiation.5 BRAF-mutated tumors occur in advanced age, females, smokers, and those with right-sided tumors.6 In CRC, most of the BRAF mutations are sporadic MSI tumors due to MLH1 promoter methylation.67 Very few mismatch repair (dMMR) tumors are due to germline mutations such as in Lynch's syndrome (LS).89 RAS-mutated CRCs were more common in males and had adenocarcinoma histology with well and moderately differentiated tumors with a microsatellite stable molecular type.10
To date, very few studies have been done on the clinical and pathological profiles of CRCs and their correlation with molecular profiles in India. The purpose of this study was to classify CRCs according to molecular subtypes and to correlate the molecular markers with the clinicopathological profile.
Aims and Objectives
To study the clinical and epidemiological profile of CRCs in a tertiary cancer care hospital in India.
To study the molecular subtypes (profile) of CRCs and their correlation with the clinicopathological profile in a tertiary cancer care hospital in India.
- To evaluate the clinical and pathological profile of CRCs.
- To determine the frequency of molecular subtypes of CRCs.
- To correlate between the molecular subtypes and their clinicopathological features.
- To determine the association between different molecular subtypes of CRCs.
Materials and Methods
A prospective noninvasive interventional study was done in patients (both inpatients and outpatients) of CRC (localized/locally advanced/metastatic) who came to our institute (Rajiv Gandhi Cancer Institute and Research Centre, Rohini, Delhi) from February 2019 to March 2020. Patients who younger than 18 years and in whom tissue was insufficient or not available for testing for at least three molecular markers out of 5 (KRAS, NRAS, BRAF, MSI, and MLH1 methylation) were excluded from this study. A sample size of 43 patients would be sufficient to detect 19.7% of incidence cases of CRC1 with assumptions, 5% level of significance, and 12% minimum allowable error. Sample size calculation was done using nMaster 2.0 software (CMC Vellore). The sample size is small due to financial constraints. We had collected data of total 50 patients for this study.
Data were collected from case sheets as per the proforma attached, which included history taking, physical examination, investigations (blood tests—complete blood count, kidney function tests, and liver function tests), imaging (colonoscopy/sigmoidoscopy, contrast-enhanced computed tomography of the abdomen, whole body positron emission tomography with computed tomography), and histopathology. Mismatch repair (dMMR) protein analysis expression was tested using immunohistochemistry (IHC; BenchMark XT, Ventana Medical Systems, Inc., Tucson, AZ, United States). Germline mutation analysis in MSH2, MLH1, PMS2, and MSH6 was performed based on the results of dMMR protein analysis. KRAS, NRAS, and BRAF V600E mutation analysis was done by reverse transcription polymerase chain reaction (PCR). Methylation of the CpG islands of MLH1 was done using pyrosequencing.
Statistical Analysis
Descriptive analysis was presented in mean ± SD or median (interquartile range) according to the distribution of data. Graphs such as bar charts, pie charts, and histograms are presented. The chi-squared and Fischer exact tests were applied for data analysis using SSPS 23.0. A p-value of less than 0.05 was considered to be statistically significant.
Ethical Implication
Informed consent was taken from patients undergoing blood tests for germline testing. Scientific committee and ethics committee approval was given for this study.
Financial Implication
The study has been funded by the Research Fund from the Rajiv Gandhi Cancer Institute and Research Centre, Delhi.
Results
Clinical and Epidemiological Characteristics
The baseline characteristics of the CRC patients are presented in Table 1. The median age at presentation was 53 years and the majority of the patients were males (54%). Tumor sites were the right colon (44%), left colon (38%), and rectum (18%). Of the 50 patients with CRC, 30 and 70% were current/former smokers and nonsmokers, respectively. Eight out of 50 patients consumed greater than 200 mL of alcohol daily. The majority of them (74%) had normal a body mass index (18–25), while only two patients had obesity (≥30). None of the patients had stage I CRC, while the majority (82%) had stage III and IV CRCs. Thirty percent and 14% of patients underwent upfront resection and neoadjuvant chemotherapy with or without radiotherapy followed by resection, respectively, while the rest never underwent resection due to either metastatic disease or progressive disease following neoadjuvant treatment.
|
No. of patients (n = 50) |
|
|---|---|
|
Age (y), median (range) |
53 (25–74) |
|
Gender |
|
|
Male |
27 (54%) |
|
Female |
23 (46%) |
|
Tumor site |
|
|
Right colon |
44% |
|
Left colon |
38% |
|
Rectum |
18% |
|
Stage (clinical and pathological) |
|
|
I |
0 (0%) |
|
II |
9 (18%) |
|
III |
14 (28%) |
|
IV |
27 (54%) |
|
Upfront obstruction/perforation |
15/50 (30%) |
|
Upfront liver metastases |
13/27 (48%) |
Pathological Characteristics
The predominant histologic subtype was classical adenocarcinoma (68%). Most of the CRCs were moderately differentiated (72%). The tumor sites were the right colon (44%), left colon (38%), and rectum (18%). Further distribution according to tumor site is depicted in Fig. 1.
Half of the patients had a proliferative type of tumor morphology as seen by colonoscopy, sigmoidoscopy, or histopathological examination in the resected specimens.
LVI and PNI were not assessed in seven patients due to biopsy from the metastatic site. Nearly half of the patients (22 of 43) had LVI, while around 40% had PNI. Both LVI and PNI were seen in 23% patients. In 17 patients, peritumoral lymphocytic infiltration could not be assessed as they were small biopsies. Intraepithelial lymphocytes were absent (0/HPF), low (<3/HPF), and high (≥3/HPF) in 60, 20, and 20% of the patients, respectively. Mild to moderate intratumoral lymphocytes (1–25%) were seen in 37 of 50 patients.
Almost half of the patients (49%, 16/33) did not have any peritumoral lymphocytes (PTLs). Of the 26 resected CRC patients, 52, 15, and 23% had low, intermediate, and high tumor budding scores (TBSs).
Molecular Characteristics
There were 40% of KRAS mutation (n =20), 0% of NRAS mutation (n = 0), 4% of BRAF mutation (n = 2), 56% of wild-type CRCs (n = 28), and 22% of deficient mismatch repair (dMMR) CRCs (n = 11), as shown in Table 2. Five patients with dMMR were also KRAS mutated.
|
Molecular marker |
No. of patients (n = 50) |
|---|---|
|
KRAS/NRAS |
|
|
Wild |
30 (60%) |
|
Mutant |
20 (40%) |
|
BRAF |
|
|
Wild |
48 (96%) |
|
Mutant |
2 (4%) |
|
dMMR |
|
|
Proficient |
39 (78%) |
|
Deficient |
11 (22%) |
|
MLH1 promoter region (n = 27) |
|
|
Methylation |
14 (52%) |
|
Unmethylation |
13 (48%) |
None of the dMMR patients had BRAF mutation. Three patients (3 out of 9) were found to be having germline deficiency in the MLH1 gene. None of the patients had any personal/past/family history of CRCs and LS-related cancers prior to their diagnosis.
Association between RAS Mutation and Clinicopathological Characteristics
In total, 30 CRCs were RAS wild-type tumors (60%). KRAS wild-type carcinomas were seen in 56% of patients older than 50 years and 63% of males, which were not statistically different from those for mutated KRAS carcinoma at 44 and 37%, respectively. RAS-mutated patients are less likely to present with upfront obstruction symptoms and advanced stages (stages III and IV) even though the numbers did not reach statistical significance. RAS-mutated tumors were associated with upfront liver metastases (69 vs. 31%; p = 0.019) and were poorly differentiated (92 vs. 50%, p = 0.020) as compared with RAS wild-type tumors, as shown in Table 3.
|
Clinicopathological characteristics |
KRAS/NRAS |
BRAF |
dMMR |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
No. of patient (%) |
Mutant, 40% |
Wild, 60% |
p-value |
No. of patients (%) |
Mutant, 4% |
Wild, 96% |
p-value |
No. of patients (%) |
Deficient, 22% |
Proficient, 78% |
p-value |
|
|
Age |
50 (100) |
0.538 |
50 (100) |
0.322 |
50 (100) |
0.728 |
||||||
|
< 50 y |
16 (32) |
5 (31) |
11 (69) |
16 (32) |
0 (0) |
16 (100) |
16 (32) |
4 (25) |
12 (75) |
|||
|
≥50 y |
34 (68) |
15 (44) |
19 (56) |
34 (68) |
2 (6) |
32 (94) |
34 (68) |
7 (20) |
27 (80) |
|||
|
Gender |
50(100) |
0.774 |
50 (100) |
0..493 |
50(100) |
0.046 |
||||||
|
Male |
27 (54) |
10 (37) |
17 (63) |
27 (54) |
2 (74) |
25 (26) |
27 (54) |
9 (33) |
18 (67) |
|||
|
Female |
23 (46) |
10 (43) |
13 (57) |
23 (46) |
0 (0) |
23 (100) |
23 (46) |
2 (9) |
21 (91) |
|||
|
Smoking status |
50 (100) |
1.000 |
50 (100) |
0.528 |
50 (100) |
0.205 |
||||||
|
Former/current smokers |
15 (30) |
6 (40) |
9 (60) |
15 (30) |
1 (67) |
14 (33) |
15 (30) |
5 (33) |
10 (67) |
|||
|
Never smokers |
35 (70) |
14 (40) |
21 (60) |
35 (70) |
1 (3) |
34 (97) |
35 (70) |
6 (17) |
29 (83) |
|||
|
Alcohol |
50 (100) |
0.875 |
50 (100) |
0.297 |
50 (100) |
0.823 |
||||||
|
Yes |
8 (16) |
3 (38) |
5 (62) |
8 (16) |
1 (12) |
7 (88) |
8 (16) |
2 (25) |
6 (75) |
|||
|
No |
42 (84) |
17 (40) |
25 (60) |
42 (84) |
1 (2) |
41 (98) |
42 (84) |
9 (21) |
33 (79) |
|||
|
BMI |
50 (100) |
0.198 |
50 (100) |
0.899 |
50 (100) |
0.722 |
||||||
|
< 18 |
2 (4) |
2(100) |
0 (0) |
2 (4) |
0 (0) |
2 |
2 (4) |
0 (0) |
2 (100) |
|||
|
18–25 |
37 (74) |
13 (35) |
24 (65) |
37 (74) |
2 (5) |
35 (95) |
37 (74) |
9 (24) |
28 (76) |
|||
|
25–30 |
8 (16) |
3 (37) |
5 (63) |
8 (16) |
0 (0) |
8 (100) |
8 (16) |
1 (12) |
7 (88) |
|||
|
> 30 |
3 (6) |
2 (67) |
1 (33) |
3 (6) |
0 (0) |
3 (100) |
3 (6) |
1 (33) |
2 (67) |
|||
|
Upfront obstruction |
50 (100) |
1.000 |
50(100) |
0.345 |
50 (100) |
0.823 |
||||||
|
Yes |
15 (30) |
6 (40) |
9 (60) |
15 (30) |
0 (0) |
15 (100) |
15 (30) |
3 (20) |
12 (80) |
|||
|
No |
35 (70) |
14 (40) |
21 (60) |
35 (70) |
2 (6) |
33 (94) |
35 (70) |
8 (23) |
27 (77) |
|||
|
Upfront liver metastases |
26 (52) |
0.019 |
26 (52) |
26 (52) |
0.000 |
|||||||
|
Yes |
13 (50) |
9 (69) |
4 (31) |
10 |
1 (10) |
9 (90) |
0.605 |
13 (50) |
0 (0) |
13 (100) |
||
|
No |
13 (50) |
2 (15) |
11 (85) |
16 |
0 (0) |
16 (100) |
13 (50) |
0 (0) |
13 (100) |
|||
|
Stage |
50 (100) |
0.454 |
50(100) |
0.499 |
50 (100) |
0.002 |
||||||
|
I + II |
9 (18) |
5 (56) |
4 (44) |
9 (18) |
0 (0) |
9 (100) |
9 (18) |
6 (67) |
3 (33) |
|||
|
III + IV |
41 (82) |
15 (36) |
26 (64) |
41 (82) |
2 (5) |
39 (95) |
41 (82) |
5 (12) |
36 (88) |
|||
|
Tumor site |
50 (100) |
0.687 |
50(100) |
0.790 |
50 (100) |
0.056 |
||||||
|
Right colon |
22 (44) |
10 (45) |
12 (55) |
22 (44) |
1 (5) |
21 (95) |
22 (44) |
8 (36) |
14 (64) |
|||
|
Left colon |
19 (38) |
6 (31) |
12 (69) |
19 (38) |
1 (5) |
18 (95) |
19 (38) |
1 (5) |
18 (95) |
|||
|
Rectum |
9 (18) |
3 (33) |
6 (67) |
9 (18) |
0 (0) |
9 (100) |
9 (18) |
2 (22) |
7 (78) |
|||
|
Clinicopathological characteristics |
KRAS/NRAS |
BRAF |
dMMR |
|||||||||
|
No. of patients (%) |
Mutant |
Wild |
p-value |
No. of patients (%) |
Mutant |
Wild |
p-value |
No. of patients (%) |
Deficient |
Proficient |
p-value |
|
|
T stage |
50 (100) |
0.904 |
50 (100) |
0.530 |
50 (100) |
0.010 |
||||||
|
T3 |
18 (36) |
7 (39) |
11 (61) |
18 (36) |
0 (0) |
18 (100) |
18 (36) |
8 (44) |
10 (56) |
|||
|
T4 |
32 (74) |
13 (40) |
19 (60) |
32 (74) |
2 (6) |
30 (94) |
32 (74) |
3 (9) |
29 (91) |
|||
|
N stage |
50 (100) |
0.454 |
50 (100) |
0.499 |
50 (100) |
0.002 |
||||||
|
N0 |
9 (18) |
5 (56) |
4 (44) |
9 (18) |
0 (0) |
9 (100) |
9 (18) |
6 (67) |
3 (33) |
|||
|
N1or N2 |
41 (82) |
15 (36) |
26 (64) |
41 (82) |
2 (5) |
39 (95) |
41 (82) |
5 (12) |
36 (88) |
|||
|
M stage |
50 (100) |
0.728 |
50 (100) |
0.954 |
50 (100) |
0.000 |
||||||
|
M0 |
24 (48) |
9 (37) |
15 (63) |
24 (48) |
1 (4) |
23 (96) |
24 (48) |
11 (46) |
13 (54) |
|||
|
M1 |
26 (52) |
11 (42) |
15 (58) |
26 (52) |
1 (4) |
25 (96) |
26 (52) |
0 (0) |
26 (100) |
|||
|
Tumor morphology |
50 (100) |
1.000 |
50 (100) |
1.000 |
50 (100) |
0.496 |
||||||
|
Ulcerative |
25 (50) |
10 (40) |
15 (60) |
25 (50) |
1 (4) |
24 (96) |
25 (50) |
4 (16) |
21 (84) |
|||
|
Proliferative |
25 (50) |
10 (40) |
15 (60) |
25 (50) |
1 (4) |
24 (96) |
25 (50) |
7 (28) |
18 (72) |
|||
|
Tumor histology |
50 (100) |
0.462 |
50 (100) |
0.612 |
50 (100) |
0.410 |
||||||
|
Classic adenocarcinoma |
34 (68) |
13 (38) |
21 (62) |
34 (68) |
2 (6) |
32 (94) |
34 (68) |
6 (18) |
28 (82) |
|||
|
Mucinous/signet cell |
15 (30) |
6 (40) |
9 (60) |
15 (30) |
0 (0) |
15 (100) |
15 (30) |
5 (33) |
10 (67) |
|||
|
Serrated adenocarcinoma |
1 (2) |
1 (100) |
0 (0) |
1 (2) |
0 (0) |
1 (100) |
1 (2) |
0 (0) |
1 (100) |
|||
|
Tumor grade |
50 (100) |
0.020 |
50 (100) |
0.020 |
50 (100) |
0.019 |
||||||
|
Well/moderately differentiated |
36 (72) |
18 (50) |
18 (50) |
36 (72) |
0 (0) |
36 (100) |
36 (72) |
11 (30) |
25 (70) |
|||
|
Poorly differentiated |
24 (28) |
2 (8) |
12 (92) |
24 (28) |
2 (8) |
12 (92) |
24 (28) |
0 (0) |
14 (100) |
|||
|
LVI |
43 (86) |
0.374 |
43 (86) |
0.323 |
43 (86) |
0.795 |
||||||
|
Present |
18 (42) |
6 (33) |
12 (67) |
22 (42) |
1 (4) |
21 (96) |
22 (52) |
6 (33) |
16 (67) |
|||
|
Absent |
25 (58) |
10 (40) |
15 (60) |
21 (58) |
0 (0) |
21 (100) |
21 (48) |
5 (20) |
16 (80) |
|||
|
PNI |
43 (86) |
0.052 |
43 (86) |
0.413 |
43 (86) |
0.728 |
||||||
|
Present |
17 (40) |
3 (18) |
14 (82) |
17 (40) |
0 (0) |
17 (100) |
17 (40) |
5 (29) |
12 (71) |
|||
|
Absent |
26 (60) |
13 (50) |
13 (50) |
26 (60)) |
1 (4) |
25 (96) |
26 (60) |
6 (23) |
20 (77) |
|||
|
LN harvest ratio |
26 (52) |
0.224 |
26 (52) |
NA |
26 (52) |
0.114 |
||||||
|
0 |
12 (46) |
6 (50) |
6 (50) |
12 (46) |
0 (0) |
12(100) |
12 (46) |
7 (58) |
5 (42) |
|||
|
0.01–0.17 |
8 (30) |
1 (12) |
7 (88) |
8 (30) |
0 (0) |
6 (100) |
8 (30) |
1 (12) |
7 (88) |
|||
|
> 0.17 |
6 (24) |
2 (33) |
4 (67) |
6 (24) |
0 (0) |
8 (100) |
6 (24) |
2 (33) |
4 (67) |
|||
|
Tumor budding score |
26 (52) |
0.898 |
26 (52) |
NA |
26 (52) |
0.483 |
||||||
|
Low |
16 (62) |
6 (37) |
10 (63) |
16 (62) |
0 (0) |
16 (100) |
16 (62) |
7 (44) |
9 (56) |
|||
|
Intermediate/high |
10 (38) |
4 (40) |
6 (60) |
10 (38) |
0 (0) |
10 (100) |
10 (38) |
3 (30) |
7 (70) |
|||
|
IEL |
50 (100) |
0.470 |
50 (100) |
0.470 |
50 (100) |
0.072 |
||||||
|
Absent/low |
40 (80) |
17 (42) |
23 (58) |
40 (80) |
2 (5) |
38 (95) |
40 (80) |
7 (17) |
34 (83) |
|||
|
High |
10 (20) |
3 (30) |
7 (70) |
10 (20) |
0 (0) |
10 (100) |
10 (20) |
4 (40) |
5 (50) |
|||
|
ITL |
50 (100) |
0.652 |
50 (100) |
0.382 |
50 (100) |
0.419 |
||||||
|
Absent |
12 (24) |
4 (33) |
8 (67) |
12 (24) |
0 (0) |
12 (100) |
12 (24) |
1 (8) |
11 (92) |
|||
|
Mild/minimal |
26 (52) |
12 (46) |
14 (54) |
26 (52) |
2 (8) |
24 (92) |
26 (52) |
7 (30) |
19 (70) |
|||
|
Moderate/high |
12 (24) |
4 (33) |
8 (67) |
12 (24) |
0 (0) |
12 (100) |
12 (24) |
3 (25) |
9 (75) |
|||
|
PTL |
33 (66) |
0.560 |
33 (66) |
NA |
33 (66) |
0.010 |
||||||
Abbreviations: BMI, Body Mass Index; BRAF, V-Raf Murine Sarcoma Viral Oncogene Homolog B, IEL, Intra-Epithelial Lymphocytes; ITL, Intra -Tumoral Lymphocytes; KRAS, Kirsten RAt sarcoma; LVI, Lympho-Vascular Invasion; M, Metastasis; NRAS, Neuroblastoma RAt Sarcoma; N, Node; PNI, Peri-Neural Invasion; PTL, Peri-Tumoral Lymphocytes; T, Tumour.
Association between BRAF Mutation Status and Clinicopathological Features
BRAF wild-type CRCs were significantly poorly differentiated than BRAF mutant-type CRCs (92 and 8%, respectively; p = 0.020).
Association between dMMR Protein Expression and Clinicopathological Features
dMMR status was done in all 50 patients. Eleven (22%) patients had dMMR. MLH1, PMS2, MSH2, and MSH6 deficiency rates were 16% (8/50), 2% (1/50), 4% (2/50), and 0% (0/50), respectively. MLH1/PMS2 and MSH2/MSH6 deficiency was seen in 18% (9/50) and 4% (2/50) of CRCs, respectively. The correlation of clinicopathological characteristics with dMMR status is presented in Table 3.
Females had a higher proportion of pMMR (91 vs. 67%, p < 0.046). pMMR status was also associated with well differentiation (p < 0.022), right-sided colonic tumors (p = 0.024), and absence of PTLs. There was a definite male preponderance, with 81% of dMMR tumors detected in males. The majority of dMMR tumors were stage II cancers (67%, p < 0.001). dMMR patients had a lower propensity to invade the bowel wall (p = 0.010), nodal metastases (p = 0.006), and distant metastases. Most dMMR tumors (72%) were located on the right side of the colon. Almost half (5/11) of the dMMR tumors were mucin producing and the rest were found to have classical adenocarcinomas.
Moreover, a high number of intraepithelial lymphocytes and mild peritumoral lymphocytic infiltration was statistically associated with dMMR (40 vs. 17%, p = 0.072 and 62 vs. 38%, p = 0.010, respectively) as compared to pMMR CRCs. As seen in Table 3, at the time of resection and biopsy, fewer dMMR tumors showed lymphovascular or perineural invasion and many had low lymph node harvest ratio (LNR) and TBS although they were not statistically significant.
Association between dMMR and RAS/BRAF Mutation Status
Five (25%) of the 20 CRCs with KRAS mutations were dMMR, whereas most of the CRCs (78%) in the other two subgroups (BRAF mutant and RAS/BRAF wild type) were dMMR proficient (Table 4). These associations were statistically insignificant (p = 0.714).
|
Molecular marker |
MLH1 promoter |
||
|---|---|---|---|
|
Methylation (n = 14) |
Unmethylation (n = 13) |
p-value |
|
|
dMMR |
|||
|
Deficient (n = 7) |
3 (43) |
4 (57) |
0.2500 |
|
Proficient (n = 20) |
11 (55) |
9 (45) |
|
|
KRAS/NRAS |
|||
|
Wild (n = 8) |
8(53) |
7(47) |
1.0000 |
|
Mutant (n = 6) |
6(50) |
6(50) |
|
|
BRAF |
|||
|
Wild (n = 27) |
14 (52) |
13 (48) |
1.0000 |
|
Mutant (n = 0) |
0 (0) |
0 (0) |
|
Association between MLH1 Methylation and dMMR, KRAS, and BRAF
MLH1 methylation data of 27 patients were analyzed. MLH1 promoter methylation at the 5′ site was seen in 14 of 27 patients (52%). Of the seven dMMR tumors, three had MLH1 methylation and the rest were nonmethylated. The majority of dMMR proficient tumors were MLH1 methylated (55%). KRAS/BRAF mutation status and MLH1 methylation had no significant association (Table 4).
Discussion
This study demonstrates that abnormalities of the KRAS gene are an important finding in colorectal neoplasia in the Indian population. The data correlate with the KRAS mutation prevalence from different countries such as United States (44%),10 Japan (33.5%),11 China (40.4%),12 and other studies from India.13 In contrast to our results, Bisht et al14 found a lower prevalence of KRAS mutation (23.5%) in Indian CRCs. Although statistically insignificant, we found the frequency of RAS mutations to be low in males and younger patients with CRC (<50 years). This is dissimilar to the findings in the Bisht et al14 study, where KRAS mutations were significantly higher in older patients and females. This variable prevalence of KRAS mutation can be attributed to genetic factors, dietary factors, environmental factors, testing method, and quality of the sample.
The KRAS-mutated tumors were mostly seen in patients with classical adenocarcinoma in western studies.10 On the contrary, we found it to be not statistically significant with adenocarcinoma histology. In our study, KRAS wild-type tumors were associated with well-differentiated tumors unlike the observation by Veldore et al.13 This dissimilarity could be due to the small sample size. Being a rare mutation,15 we could also not find any NRAS mutation in our study.
BRAF mutation frequency was found to be 4% in this study, which is higher than the reported frequencies in China (2.3%) and lower than that reported in India (9.8%). The small sample size and different sensitivities for the molecular techniques used can explain the disparity. In our study, BRAF wild-type tumors were more commonly associated with well differentiation. These findings are inconsistent with the study by Bisht et al.14 As in the Li et al16 study, there was no significant difference with BRAFV600E-mutated tumors according to age and sex. Furthermore, in line with prior studies and literature, none of the KRAS mutation cases had concomitant BRAF mutations, which indicates the mutually exclusive nature of these mutations. KRAS- and BRAFV600E-mutated tumors were more advanced tumors with ≥4 positive lymph nodes and higher TNM stages. However, this observation was not consistent with our results. The smaller size of the BRAFV600E mutation and KRAS mutation subgroups can explain this variation.
The 22% of CRCs with dMMR was in accordance with data from an Indian study (29%).17 Reports from other Asian countries showed only 10% MSI-H CRCs.518 This discrepancy can be due to the different molecular tests used and their sensitivities to some extent. Compared with PCR-based MSI testing, IHC is easy to perform and the turnaround time is very less. Most importantly, IHC helps in picking up and may detect few dMMR cases that may have been missed by PCR-based MSI testing.19 Correlations between dMMR status and clinicopathological features were contrary with prior studies. It might be due to the different inclusion criteria, environmental factors, and the variable specificity and sensitivity of the different tests. Several studies showed a significant association of dMMR colorectal tumors with a lower TNM stage, poor differentiation, high intraepithelial lymphocytes, the presence of several intratumoral lymphocytic, and PTLs, and N0 nodal stage. We could not find any association between dMMR expression and LNR. This is in contrary to the study by Berg et al20 where the authors found a significant association between MSI-H status and adequate lymph node harvest (>12).
MLH1 loss accounted for the majority dMMR CRCs (72%) and 40% of this deficiency (2 out of 5 tumors with MLH1 loss) was caused by MLH1 promoter methylation, which separates sporadic dMMR CRCs from germline mutation LS cases. This is dissimilar to Hampel et al's21 study in which around 70% were sporadic tumors. Another curious finding in our study is that the family history is deceptive and misleading.21 Therefore, all newly diagnosed CRC patients should be screened for LS using an IHC-based algorithm22 (Fig. 2) rather than on family history. In our study, dMMR/KRAS mutation, dMMR/KRAS wild-type, pMMR/KRAS mutation, and pMMR/KRAS wild-type tumors were 10, 12, 30, and 48%, respectively, which is similar to the that reported by Ye et al.23 This tumor subgrouping according to molecular subtypes is prognostic as MSS/KRAS mutant tumors had the worst survival.24 Therefore, dMMR and KRAS markers will be key for the development of a molecular prognostic scoring system for CRC in the future.
This study has its limitations. There are missing histopathological and molecular data for few parameters in view of small biopsies or biopsies from metastatic sites. A relatively small sample size could have under- or overestimated the significance of the association between the molecular markers and the clinicopathological characteristics. Nevertheless, aside from the TBS, lymph node ratio, PTLs, and MLH1 methylation, the rest of the collected data were accurate to around 95% with less than 5% of missed information. Another drawback is the consideration of molecular testing in all CRC patients. However, the low incidence of CRCs in India and the short time period available for completion of this study could explain this. On the other hand, the strength of our study was prospective data collection and statistical correlation of data from one of the cohorts of CRCs (n = 50). Above all, this is the first Indian study to correlate complete RAS and BRAF analysis, dMMR status, and MLH1 methylation with CRCs' clinicopathological features and also the association between different molecular subtypes. It helped pick up additional cases for germline testing for LS.
Conclusion
In conclusion, 40, 4, and approximately one-quarter (22%) of the colorectal tumors were KRAS mutant, BRAF mutant, and dMMR, respectively. In particular, KRAS, BRAF mutation status, dMMR expression, and MLH1 methylation have unique clinical, pathological, and molecular characteristics, which must be kept in mind when assessing in clinical trials the prognosis values of different molecular markers in CRCs. Further studies including larger cohorts of CRC patients should be done to confirm these associations.
Acknowledgments
We are very grateful and convey our sincere gratitude to Dr. Sadaf (Department of Pathology); Mr. Dushyant, Mr. Sanjeev Kumar, Mr. Surinder, Mrs. Mamta, and Mrs. Mumtaz (Department of Molecular Oncology); Mrs. Sangeeta, Mr. Arvind, and Mr. Ajay (Department of Histopathology); and Dr. Kavitha Kushwaha and Sister Renu for their constant help in performing these molecular tests and all the nursing staff of our institute for helping in coordination of these tests and reports. We thank Dr. Swarnima (Department of Research) for the constant support and guidance for funding this study and Mrs. Neelima (Librarian) for providing us with and giving access to the necessary academic material.
References
- Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(06):394-424.
- [Google Scholar]
- Assessment of colorectal cancer molecular features along bowel subsites challenges the conception of distinct dichotomy of proximal versus distal colorectum. Gut. 2012;61(06):847-854.
- [Google Scholar]
- MLH1 promoter methylation frequency in colorectal cancer patients and related clinicopathological and molecular features. PLoS One. 2013;8(03):e59064.
- [Google Scholar]
- Deficient mismatch repair and RAS mutation in colorectal carcinoma patients: a retrospective study in Eastern China. PeerJ. 2018;6:e4341.
- [Google Scholar]
- Colorectal cancer with mutation in BRAF, KRAS, and wild-type with respect to both oncogenes showing different patterns of DNA methylation. J Clin Oncol. 2004;22(22):4584-4594.
- [Google Scholar]
- Mutations of BRAF are associated with extensive hMLH1 promoter methylation in sporadic colorectal carcinomas. Int J Cancer. 2004;108(02):237-242.
- [Google Scholar]
- BRAF screening as a low-cost effective strategy for simplifying HNPCC genetic testing. J Med Genet. 2004;41(09):664-668.
- [Google Scholar]
- Promoter hypermethylation frequency and BRAF mutations distinguish hereditary non-polyposis colon cancer from sporadic MSI-H colon cancer. Fam Cancer. 2004;3(02):101-107.
- [Google Scholar]
- KRAS testing, tumor location, and survival in patients with stage IV colorectal cancer: SEER, 2010–2013. J Natl Compr Canc Netw. 2017;15(12):1484-1493.
- [Google Scholar]
- Prognostic relevance of KRAS and BRAF mutations in Japanese patients with colorectal cancer. Int J Clin Oncol. 2013;18(06):1042-1048.
- [Google Scholar]
- KRAS and BRAF gene mutations in correlation with clinicopathologic features of colorectal carcinoma in Chinese. Zhonghua Bing Li Xue Za Zhi. 2012;41(09):584-589.
- [Google Scholar]
- Prevalence of KRAS mutations in metastatic colorectal cancer: a retrospective observational study from India. Indian J Cancer. 2014;51(04):531-537.
- [Google Scholar]
- Molecular spectrum of KRAS, BRAF, and PIK3CA gene mutation: determination of frequency, distribution pattern in Indian colorectal carcinoma. Med Oncol. 2014;31(09):124.
- [Google Scholar]
- NRAS mutations are rare in colorectal cancer. Diagn Mol Pathol. 2010;19(03):157-163.
- [Google Scholar]
- Colorectal carcinomas with KRAS codon 12 mutation are associated with more advanced tumor stages. BMC Cancer. 2015;15:340.
- [Google Scholar]
- Prevalence estimation of microsatellite instability in colorectal cancers using tissue microarray based methods: a tertiary care center experience. Indian J Pathol Microbiol. 2018;61(04):520-525.
- [Google Scholar]
- Association of smoking, CpG island methylator phenotype, and V600E BRAF mutations in colon cancer. J Natl Cancer Inst. 2006;98(23):1731-1738.
- [Google Scholar]
- Immunohistochemistry versus microsatellite instability testing for screening colorectal cancer patients at risk for hereditary nonpolyposis colorectal cancer syndrome. Part I. The utility of immunohistochemistry. J Mol Diagn. 2008;10(04):293-300.
- [Google Scholar]
- DNA sequence profiles of the colorectal cancer critical gene set KRAS-BRAF-PIK3CA-PTEN-TP53 related to age at disease onset. PloS One. 2010;5(11):e13978.
- [Google Scholar]
- Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer) N Engl J Med. 2005;352(18):1851-1860.
- [Google Scholar]
- Prevalence of Lynch syndrome and Lynch-like syndrome among patients with colorectal cancer in a Japanese hospital-based population. Jpn J Clin Oncol. 2017;47(02):10.
- [Google Scholar]
- KRAS and BRAF gene mutations and DNA mismatch repair status in Chinese colorectal carcinoma patients. World J Gastroenterol. 2015;21(05):1595-1605.
- [Google Scholar]
- KRAS mutation and microsatellite instability: two genetic markers of early tumor development that influence the prognosis of colorectal cancer. Ann Surg Oncol. 2010;17(02):416-424.
- [Google Scholar]