Assessment of new pathological markers in early stage colon cancer: Insights and limitations

Abstract

BACKGROUND

The decision to administer adjuvant chemotherapy to patients with local stage depends on specific high-risk features that are T4 tumor stage, presence of perineural invasion, lymphovascular invasion, poorly differentiated tumor histology, inadequate lymph node sampling (fewer than 12 lymph nodes), and evidence of tumor perforation or obstruction. Tumor-stroma ratio, tumor infiltrating lymphocytes (TIL), Crohn-like reaction (CLR), desmoid reaction, poorly differentiated clusters (PDC) are new pathological markers that are being studied.

AIM

To examine the relationship between new pathological markers and defined high risk factors, in early stage colorectal cancer.

METHODS

We evaluated 155 patients with the diagnosis stage I and II colorectal cancer between the years 2007 and 2021 who were treated at Trakya University Hospital, Department of Medical Oncology. We divided those with and without high-risk factors into two groups. We examined the relationship of new pathological markers with these groups and with pathological markers in risk factors.

RESULTS

There was no statistically significant correlation between presence of TIL, presence of PDC, presence of tumor budding, presence of CLR, presence of desmoid reaction and low and high-risk groups according to the degree of those with PDC (P = 0.82, P = 0.51, P = 0.77, P = 0.37, P = 0.83, respectively). In addition, no statistically significant correlation was found between the tumor-stroma ratio and low and high risk groups (P = 0.80). We found a statistically significant correlation between the presence of PDC and the presence of PDC grade 3 and T stage (P = 0.001, P = 0.001, respectively). It was determined that the presence of PDC and the frequency of grade 3 PDC increased with the advanced T stage.

CONCLUSION

No relationship was found between the presence of new pathological markers and high-low risk groups. When we examined the relationship between new and old pathological markers, only the frequency of detection of PDC and PDC grade 3 was found to be correlated with advanced T stage.

Key Words: Colon cancer; Early stage; Risk factors; New pathological markers; Tumor-stroma ratio; Tumor infiltrating lymphocytes; Crohn-like reaction; Desmoid reaction; Poorly differentiated clusters

Core Tip: When their associations with the pathological marker in the risk group were evaluated, it was found that there was a significant correlation between the presence of poorly differentiated clusters (PDC) and the presence of PDC grade 3 and advanced T stage. This showed us that PDC is one of the markers to be used in cases where these risk factors are insufficient.

INTRODUCTION

Colorectal cancer (CRC) is the third most common cancer worldwide and one of the leading causes of cancer-related deaths worldwide[1]. It accounts for about 10% of all cancers diagnosed alone, and about 35% consists of stage I and II patients[2]. Although about one third of colon cancer is diagnosed at local stage, specific global epidemiological data for stage II colon cancer are limited. On the other hand, within about 5 years, 25%-30% of patients with localized colon cancer progress to metastatic stage. However, the decision to administer adjuvant chemotherapy is nuanced and depends on specific high-risk features adjuvant chemotherapy is the standard treatment approach for patients with high-risk stage II disease; however, it is not routinely recommended for those with low-risk stage II disease. Furthermore, classical pathological criteria for high-risk designation include T4 tumor stage, presence of perineural invasion (PNI), lymphovascular invasion (LVI), poorly differentiated tumor histology, inadequate lymph node sampling (fewer than 12 lymph nodes), and evidence of tumor perforation or obstruction[3].

The prognosis for patients with stage II CRC (T3-T4N0M0) is heterogeneous, with 5-year overall survival rates ranging from 87.5% in stage IIA to 58.4% in stage IIC. These survival rates may vary internationally due to differences in access to early screening programs. Postoperative chemotherapy is believed to provide greater benefits for patients at high risk of recurrence than for patients at low risk[4]. Adjuvant therapy is recommended by guidelines only for “high-risk” stage II CRC patients[5]. Moreover, a significant unmet need exists for validated predictive biomarkers to accurately identify stage II CRC patients who would benefit from adjuvant therapy. Advances in the molecular characterization of CRC are anticipated to enhance the identification of stage II patients who may gain greater benefit from adjuvant chemotherapy or achieve higher cure rates with surgery alone. In addition, standard adjuvant chemotherapy improves the 5-year survival rate by only 2%-5% in patients with stage colon cancer[6]. New pathological and molecular markers, such as poorly differentiated clusters (PDC), tumor budding (TB), circulating tumor DNA, and immune score, have demonstrated significance in predicting prognosis and recurrence[2]. However, they have not yet been defined as risk factors that will guide treatment.

In this study, we aimed to evaluate the associations between novel pathological markers-including tumor-stroma ratio (TSR), tumor-infiltrating lymphocytes (TIL), Crohn-like reaction (CLR), desmoid reaction (DR), TB, and PDC and established high-risk factors in patients with stage I and II CRC.

MATERIALS AND METHODS

Study design

This study was conducted with a retrospective study design. Medical records of 218 non-metastatic colon cancer patients were initially evaluated. Of these, 155 patients with complete data were included in the final analysis. The medical records of resected stage I and stage II CRC patients admitted to oncology outpatient clinics followed up regularly between 2007 and 2021 were retrospectively analyzed. Patients with stage 3 or metastatic disease, rectal cancer, a second primary malignancy, whom paraffin embedded tissue samples were not present in the pathology archive and insufficient data in the follow-up file were excluded. Patients older than 18 years of age, with stage I and II disease, whose pathology samples were available, were included in the study. All baseline data of the patient demographics, disease characteristics and follow-up data were derived from archive files.

High-risk and low-risk groups were defined in accordance with criteria established by the European Society for Medical Oncology guidelines. High-risk classification was based on the presence of specific pathological features, including T4 tumor stage, PNI, LVI, poorly differentiated tumor histology, inadequate lymph node dissection (fewer than 12 nodes), and evidence of tumor perforation or obstruction. Patients lacking these high-risk characteristics were classified as low-risk[3]. We examined the relationship between the presence of PDC, PDC grade, TSR, presence of CLR, presence of TB, presence of TIL, presence of DR and risk groups, and the relationship between pathological findings (for example, LVI) in risk factors. We also examined the relationship between new pathological markers and demographic data and other tumoral conditions. We also examined the relationship between the new pathological markers and the pathological marker found in high risk factors.

Histologic evaluation

All samples were processed in according the recommendations of the College of American Pathologists[7]. All available hematoxylin and eosin (HE) stained slides with full-thickness sections of the tumor, including the deepest portion of the invasive tumor (mean, 7 tumor slides per patient) were examined. The pathologist was blinded to other risk factors. A BX51 microscope (Olympus) which had a standard 22 mm diameter eyepiece (specimen area of 0.950 mm² under an objective lens with a magnification of × 20) was used.

Clusters of more than 5 cancer cells that lacked a gland-like structure were defined as PDC. First of all, the whole tumor slides, to identify areas with the highest number of PDCs, were scanned under low-power magnification. Under an objective lens with a magnification of × 20 the number of PDCs in a single hot point was identified and graded as G1 (< 5 clusters), G2 (5-9 clusters), or G3 (≥ 10 clusters)[8]. TB was defined as a single cancer cell or cluster of less than 5 cancer cells on the invasive front and evaluated as present or absent. was defined as Nodular lymphoid aggregates of 1 mm or larger at the tumor periphery was defined as CLR and CLR classified as present or absent[9]. TIL positivity defined as presence of lymphocytes within the boundaries of tumor cell nests or glands. Based on the proportion of gland formation World Health Organization tumor grade was categorized as grade 1 (well-differentiated, > 95%), grade 2 (moderately differentiated, 50%-95%) or grade 3 (poorly differentiated, 0%-49%)[10]. In the reactive fibrous zone DR (desmoplastic reaction) was evaluated and classified as present or absent[9]. Mismatch repair protein deficiency testing was performed with standard streptavidin biotin-peroxidase method.

On routine 5 μm thick HE sections TSR is assessed with conventional microscopy. In the invasive part of the tumor, which is most decisive for tumor progression intratumoral stroma formation is evaluated. The tumor usually corresponds to the slide used to determine the T status in routine pathology and is noted in the pathology report. Areas covered with the greatest amount of stroma are selected using the 92.5 or 95 target. Using the 910 Lens, image areas are scored in 10% increments. Identifying a single image area with high stroma content is decisive for a final stroma classification[11]. We expressed the measurements in percentiles for each of the tumor and stroma tissue separately.

Statistical analysis

Statistical analyses were performed using statistical product and service solutions software, version 22. Missing data were managed by excluding cases with incomplete information from the relevant analyses. Data were presented as median and 25^th-75^th interquartile range. Categorical variables were reported as frequencies and group percentages. Demographic, clinical features with new pathologic findings in stage I and stage II tumors were compared. Non-parametric variables were studied by Ki-Square test. Parametric variables were compared with independent-sample t-test. Multivariate analysis could not be done there was less recurrences as expected in early stage colon cancer. A P value less than 0.05 was considered as statistically significant.

RESULTS

A total of 155 patients, 67 (43.5%) women and 88 (56.8%) men, were included in the study. The mean age was 67 (range 60-74). Right colon tumors were detected in 58 (37.6%) patients and left colon tumors in 97 (62.4%) patients. There was a mucinous component in the pathology of 32 (20.6%) patients. Tumors of 144 (92.9%) patients were grade 1 and 2, and tumors of 11 (7.9%) patients were grade 3. Obstruction or perforation was detected in the pathology of 3 (1.9%) patients. In the pathology of 35 (22.6%) patients, LVI was positive. PNI was positive in 14 (9%) patients. Pathologically, 7 (4.5%) patients were T1, 37 (23.9%) patients T2, 88 (56.8%) T3, 23 (14.8%) patients T4. There were 102 (65.8%) patients whose < 12 lymph nodes were dissected. There were 25 (16.1%) patients with intact microsatellite instability (MSI) and 12 (7.7%) patients with MSI-high. The MSI status of the other patients was unknown. Considering the risk factors, 25 (16.1%) patients were included in the low-risk group and 130 (83.9%) patients were included in the high-risk group. TIL was detected in 105 (67.7%) patients. PDC was detected in 91(58.7%) patients. Grade 1 PDC was detected in 43 (47.2%) of these 91 patients, grade 2 in 27 (29.7%) and grade 3 in 21 (23.1%). There were 58 (37.4%) TB positives. CLR was positive in 21 (13.5%) patients and DR was positive in 86 (55.5%) patients. TSR was found to be ≤ 1 (low) in 36 (23.2%) patients and > 1 (high) in 119 (76.8%) patients (Table 1).

Table 1 Demographic data, risk factors, pathological markers, other tumoral conditions.

Characterization	Number	Percentage rates (%)
Age (years)
< 67	77	49.7
≥ 67	78	50.3
Gender
Female	67	43.2
Male	88	56.8
Tumor location
Right colon	58	37.4
Left colon	97	62.6
Pathology
Mucinous	32	20.6
Nonmucinous	123	79.4
Tumor grade
Grade 1-2	144	92.9
Grade 3	11	7.1
Presence of perforation or obstruction
Yes	3	1.9
No	152	98.1
Surgical margin
Positive	1	0.6
Negative	154	99.4
Lymphovascular invasion
Positive	35	22.6
Negative	120	77.4
Perineural invasion
Positive	14	9.1
Negative	141	90.9
T stage
T1	7	4.5
T2	37	23.9
T3	88	56.8
T4	23	14.8
Lymph node dissection
< 12	102	65.8
≥ 12	53	35.2
MSI status
Intact	25	16.1
High	12	7.7
Unknown	118	76.1
Risk group
Low risk	25	16.1
High risk	130	83.9
TIL presence
Positive	50	32.3
Negative	105	67.7
Poorly differentiated clusters
Positive	91	58.7
Negative	64	41.3

MSI: Microsatellite instability; TIL: Tumor infiltrating lymphocytes.

No statistically significant correlation was found between tumor location, presence of mucinous component, age and gender regardless of its rate, and low and high risk groups. There was no statistically significant correlation between presence of TIL, presence of PDC, presence of TB, presence of CLR, presence of DR and low and high-risk groups according to the degree of those with PDC (P = 0.82, P = 0.51, P = 0.77, P = 0.37, P = 0.83, respectively). In addition, no statistically significant correlation was found between the TSR and low and high risk groups (P = 0.80) (Table 2).

Table 2 Relationship of data with risk groups, n (%).

Characterization	Low risk group (n = 25)	High risk group (n = 130)	P value
Age (years)			0.83
< 67	13 (52)	64 (49.2)
≥ 67	12 (48)	66 (50.8)
Gender			0.66
Female	12 (48)	55 (42.3)
Male	13 (52)	75 (56.7)
Tumor location			0.12
Right colon	13 (52)	45 (34.6)
Left colon	12 (48)	85 (65.4)
Pathology			0.17
Mucinous	8 (32)	24 (18.5)
Nonmucinous	17 (68)	106 (81.5)
TIL presence			0.82
Positive	7 (28)	43 (33.1)
Negative	18 (72)	87 (66.9)
PDC			0.51
Positive	13 (52)	78 (60)
Negative	12 (48)	52 (40)
PDC grade (n = 91)			0.99
Grade 1	6 (46.1)	37 (47.4)
Grade 2	3 (23.1)	24 (30.8)
Grade 3	4 (30.8)	17 (21.8)
Tumor budding			0.37
Positive	7 (28)	51 (39.2)
Negative	18 (72)	79 (60.8)
Crohn-like reaction			0.99
Positive	3 (12)	18 (13.8)
Negative	22 (88)	112 (86.2)
Desmoid reaction			0.83
Positive	13 (52)	73 (56.2)
Negative	12 (48)	57 (43.8)
Tumor-stroma ratio			0.80
Low	5 (20)	31 (23.8)
High	20 (80)	99 (76.2)

PDC: Poorly differentiated clusters; TIL: Tumor infiltrating lymphocytes.

When other results were evaluated, it was found that demographic data and other tumoral conditions were similarly distributed between patients with high and low risk factors. In addition, the frequency of detection of new pathological markers was found to be similar between patients with high or low risk factors.

When examining the interaction between pathological findings, a statistically significant correlation was found between the presence of PDC and the presence of PDC grade 3 and T stage (P = 0.001, P = 0.001, respectively). Based on this result, it was determined that the presence of PDC and the frequency of grade 3 PDC increased with the advanced T stage.

DISCUSSION

This study evaluated the distribution and significance of new pathological markers and risk factors in patients with Stage II colon cancer. The findings revealed a strong association between the presence of PDC and grade 3 PDC with advanced T stage. This suggests that PDC may serve as an important indicator in identifying advanced-stage disease and should be considered in clinical management. However, other new pathological markers, including TIL, TB, CLR, DR, and TSR, did not show a significant correlation with risk groups. These results indicate that PDC may have a stronger prognostic value compared to other markers and warrant further investigation in future studies.

Stage II colon cancer is a highly heterogeneous disease. Despite the absence of lymph node involvement, certain pathological and clinical features can indicate a higher risk of recurrence and poorer prognosis. Among these features, the presence of poorly differentiated tumor cells has been identified as one of the strongest clinicopathological indicators for predicting lymph node metastasis in submucosal invasive colon cancer[12,13]. Our study reinforces these findings by demonstrating a statistically significant relationship between the presence of PDC, the grade of PDC, and advanced T stage. Specifically, as the T stage progresses, the prevalence of PDC increases, and the likelihood of encountering grade 3 PDC becomes higher. Moreover, the findings emphasize the necessity for a deeper understanding of the tumor microenvironment and its role in cancer progression. The differential impact of PDC compared to other markers raises questions about the biological mechanisms underlying its association with advanced disease stages. This correlation suggests that PDC could serve as a critical marker for identifying patients at higher risk of disease progression, thus influencing clinical management decisions.

TIL play a significant role in the classification and prognosis of stage II colon cancer. By comparing TIL levels in high and low-risk groups, clinicians can refine their risk stratification and tailor treatment strategies more effectively. High TIL levels generally correlate with better outcomes, making them a valuable marker in predicting patient prognosis. Moreover, TILs are immune cells that have migrated into the tumor microenvironment. Their presence has been widely studied as a prognostic marker in various cancers, including stage II colon cancer. TILs are indicative of the host immune response against the tumor and are generally considered a favorable prognostic factor. In addition, many studies have found a positive relationship between the presence of TIL and overall survival[14,15]. In a meta-analysis, it is stated that the presence of TIL provides a prognostic advantage for overall survival with hazard ratio = 0.65[16]. In addition, in high-risk patients, the presence of TILs can provide additional prognostic information. Even within this group, those with higher levels of TILs might have a relatively better prognosis compared to those with low TIL levels, suggesting that TILs can modify the risk stratification within this subgroup. In our study, it was observed that there was no any association between TIL frequency and the risk. This lack of association suggests that, in this cohort, TILs may not serve as a reliable marker for risk stratification within stage I and II CRC. Future studies may further investigate the conditions under which TILs could refine risk assessment in similar patient populations.

TB is defined by the presence of single tumor cells or small clusters of cells within the tumor center and may reflects the epithelial-mesenchymal transition, a process by which epithelial cells acquire mesenchymal, invasive properties, facilitating tumor dissemination and metastasis. It is considered a marker of tumor aggressiveness and has been widely studied in various stages of CRC, including stage II colon cancer. In addition, it was demonstrated that TB grades and relapse-free time were statistically associated and grade 3 TB was more common in T4 individuals and those with positive LVI[17]. In our study, TB was evaluated based on its presence or absence. However, this assessment did not reveal any significant associations with other pathological parameters.

DR is a potentially valuable marker in the classification and prognosis of stage II colon cancer. The assessment of DR can aid in refining risk stratification for stage II colon cancer patients. While its presence may suggest a more controlled tumor environment and better prognosis, its exact role in high-risk and low-risk groups is unknown. Previous studies have categorized DR into three distinct types: Immature, intermediate, and mature. Notably, a statistically significant association was observed between the presence of immature DR and recurrence in stage II colon cancer[18]. In contrast, our analysis assessed DR solely based on its presence or absence, and we did not find any significant correlation between DR frequency and other risk factors or pathological findings.

CLR is characterized by the presence of nodular lymphoid aggregates or granulomas in the peritumoral stroma, resembling the inflammatory response seen in Crohn’s disease. CLR represents an immune response against tumor and can be indicative of an active host immune defense. Moreover, CLR has been associated with a lower incidence of local recurrence, fewer distant metastases, and better cancer-specific and overall survival[14,19,20]. We found no association between CLR frequency and abnormal results in risk factor groups or risk factors. On the other hand, MSI status can provide a more comprehensive understanding of the tumor’s immune landscape. Similarly, MSI-high tumor microenvironment includes higher peritumoral lymphocyte infiltration. Assessing both CLR and MSI status can provide a more comprehensive understanding of the tumor’s immune landscape, aiding in risk stratification as well. In our study, we did not find any association between the frequency of CLR and abnormal results in risk factor groups or other pathological risk factors. Additionally, no statistically significant correlation was observed between CLR and MSI status, though the MSI subgroup was small.

TSR refers to the proportion of stromal tissue (connective tissue) relative to the tumor tissue within the tumor microenvironment. It is typically evaluated on routine HE stained sections of the tumor, focusing on the invasive front. Moreover, a dense stromal environment can support tumor growth and metastasis through various mechanisms such as enhanced angiogenesis and immune evasion. The TSR is a significant prognostic marker in stage II colon cancer. Its assessment can aid in better risk stratification. In addition, patients with tumors with a high stromal content have a significantly worse overall survival and disease-free survival[20,21]. Because outcomes for patients with stage II colon cancer are highly variable, TSR is a useful tool for selecting patients at risk of disease recurrence or metastasis[13]. In our study, TSR did not show any association with the presence of risk factors or other pathological findings.

There are several limitations must be acknowledged. Firstly, the retrospective nature of the study design inherently introduces potential biases, such as selection bias and information bias, which could affect the validity and generalizability of the results. Secondly, the assessment of pathological markers, such as TB, DR, and CLR, was based on their presence or absence rather than a more granular evaluation of their grades or intensity. This simplified approach might have overlooked subtle yet clinically relevant differences. Additionally, inter-observer variability in the interpretation of histological samples could have influenced the consistency of the results. The lack of correlation between some markers and risk factors, such as TSR, CLR, and DR, could be attributed to the methodological constraints and the heterogeneity within the tumor microenvironment that were not fully captured. Despite the limitations outlined, our study presents several notable findings. The study offers valuable insights into the role of PDC and underscores the necessity for a deeper understanding of tumor microenvironment interactions in Stage II colon cancer.

CONCLUSION

In conclusion, when their associations with the pathological marker in the risk group were evaluated, it was found that there was a significant correlation between the presence of PDC and the presence of PDC grade 3 and advanced T stage. This showed us that PDC is one of the markers to be used in cases where these risk factors are insufficient.