Table of Contents    
Year : 2019  |  Volume : 10  |  Issue : 3  |  Page : 68-72  

Determination of cancer-associated fibroblast and stromal phenotypes as novel prognostic factors for colorectal carcinomas associated with tumor budding

1 Doctoral Program in Biomedical Sciences, Faculty of Medicine, Universitas Indonesia, Jakarta; Department of Anatomic Pathology, Faculty Medicine, Universitas Andalas, Padang, Indonesia
2 Department of Anatomic Pathology, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia
3 Department of Internal Medicine, Division of Gastroenterology, Faculty Medicine, Universitas Indonesia, Jakarta, Indonesia
4 Department of Biochemistry and Molecular Biology, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia

Date of Web Publication14-Jan-2020

Correspondence Address:
Septelia Inawati Wanandi
Department of Biochemistry and Molecular Biology, Faculty Medicine, Universitas Indonesia, Jakarta
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jnsbm.JNSBM_61_19

Rights and Permissions

Objective: Tumor microenvironments consisting of stroma and extracellular matrix play important roles in tumor progression, particularly through the induction of epithelial-mesenchymal transition (EMT). It has been suggested that the phenotype of cancer-associated fibroblasts (CAFs) in stroma might be correlated with the prognosis of patients with colorectal carcinoma (CRC). We aimed to determine the stromal and CAF types for prognostic determinants of CRCs associated with tumor budding (TB) grade, reflecting EMT. Materials and Methods: Using hematoxylin and eosin-stained paraffin wax sections from 23 patients with CRC, three stromal and two CAF phenotypes were evaluated, TB grades, invasion depth, lymph node metastases, and lymphovascular invasion (LVI) were also analyzed as established prognostic determinants. Data were analyzed statistically using the Chi-squared tests. Results: There was a significant association between CAF phenotype and TB grade (P < 0.01). CRC specimens with immature CAF have higher TB grades than the mature phenotype. Nevertheless, a significant association between stroma and TB grade could not be demonstrated. Moreover, high TB grades were significantly associated with lymph node metastasis (P < 0.01) and LVI (P < 0.01). However, there were no significant associations between CAF phenotype and either of these prognostic determinants. Conclusion: CAF phenotype could be considered as a prognostic determinant of CRC through its association with TB grade, indicating the role of CAFs in EMT processes. Future studies are required to examine the secretomes of CAFs that play important roles in EMT and determine the prognosis for patients with CRCs.

Keywords: Cancer-associated fibroblasts, prognostic determinants, stromal phenotype, tumor budding

How to cite this article:
Hilbertina N, Siregar NC, Abdullah M, Wanandi SI. Determination of cancer-associated fibroblast and stromal phenotypes as novel prognostic factors for colorectal carcinomas associated with tumor budding. J Nat Sc Biol Med 2019;10, Suppl S1:68-72

How to cite this URL:
Hilbertina N, Siregar NC, Abdullah M, Wanandi SI. Determination of cancer-associated fibroblast and stromal phenotypes as novel prognostic factors for colorectal carcinomas associated with tumor budding. J Nat Sc Biol Med [serial online] 2019 [cited 2020 Apr 3];10, Suppl S1:68-72. Available from:

   Introduction Top

Colorectal carcinoma (CRC) is an epithelial malignancy of the colon and rectum. In 2018, 1.8 million new cases (10.2% of all cancers) of CRC were estimated to occur worldwide, indicating that CRC is the third most commonly diagnosed cancer.[1] Although the incidence of CRC has decreased in developed countries, it is still increasing in developing countries because of the use of sedentary lifestyles, such as changes to diet, obesity, and heavy smoking.[2] Until now, the gold standard used to predict CRC prognosis is based on the tumor–node–metastasis (TNM) system established by the Union for International Cancer Control.[3] However, it has been reported that a number of low-risk CRCs defined by the TNM system progressed to local or distal recurrences [4] and tumor stages could not accurately determine patient survival for the same stage.[5] The failure of TNM to serve as a reliable prognostic system, especially for those patients with intermediate-stage tumors, could be overcome by considering morphological, molecular, or treatment-related factors that might better differentiate the diagnosis.[6] One of the promising histopathological parameters in CRCs is the presence of tumor budding (TB) which is defined as single cells or clusters of up to four cells at the invasive margin of a CRC.[3],[5] Pathologists have been paying increasing attention to TB as a prognostic factor because it indicates an aggressive tumor driven by epithelial-mesenchymal transition (EMT) processes. Moreover, the rate of TB has been accepted as a well-established independent prognostic factor for patients with CRC.[5]

Tumor microenvironments consisting of stroma and extracellular matrix (ECM) have major influences on tumor behavior, including modifying invasiveness and metastatic potential through the induction of EMT. Cancer-associated fibroblasts (CAFs) are major components of stromal cells in the tumor microenvironment that secrete growth factors, cytokines, and other ECM proteins, which are involved in the proinvasive behavior of cancer cells.[7],[8] Thus, the phenotypes of CAFs might be correlated with the prognosis for patients with CRC. CAFs and stromal phenotypes are frequently observed in daily histopathological CRC examinations, but are rarely reported. It has been reported that an immature stromal type was associated with the lowest 5-year disease-free survival rate of patients with CRC compared with mature and intermediate-stage stromas.[7] However, little is known about the association of CAFs and stromal phenotype with EMT, which determines the prognosis for patients with CRCs. The aim of this study was to determine the phenotypes of CAFs and stromas as novel independent prognostic factors for CRCs. We also evaluated the association of TB grade with other pathological prognostic factors, such as tumor invasion depth, lymph nodes metastasis, and lymphovascular invasion (LVI), which represent the EMT process in CRCs.

   Materials and Methods Top

Patients and tissues

Twenty-three patients diagnosed at the anatomical pathology laboratory of Cipto Mangunkusumo Hospital, Jakarta with colorectal adenocarcinomas, which had invaded beyond the mucosal layer of the colorectal wall, underwent surgical resection from January 2017 to July 2018. No patients were treated with neoadjuvant chemotherapy or radiotherapy. Adenocarcinoma is the most common histology subtype found in the CRC. Specimens with no lymph node samples available or for which hematoxylin and eosin (H and E)-stained slides could not be reviewed were excluded from the study. The surgically resected specimens were fixed in 10% neutral buffered formalin, embedded in paraffin wax, and stained with H and E for histopathology. The use of formalin-fixed paraffin wax-embedded (FFPE) specimens was approved by the Head of the Anatomic Pathology Department, Medicine Faculty, University of Indonesia (No. 214/UN2.F1.D/PA/PPM.00.02/2017 and No. 01/UN2.F1.D/PA/PPM.SIPP. 00.02/2017) and also by the Research Unit of Cipto Mangunkusumo Hospital (No. LB.02.01/X.2/648/2017). The research protocol was approved by the Ethics Committee of the Faculty of Medicine, University of Indonesia (No. 433/UN2.F1/ETIK/2017).


The clinicopathological parameters including patient age, gender, tumor differentiation, invasion depth (pT), lymph node metastasis (pN), LVI, and pathology diagnoses made in routine practice recorded in the pathology reports were used for analysis. Stromal types were classified histologically as mature, intermediate, or immature based on the presence of keloid-like collagen and myxoid stroma in the invasive front of the tumor. Keloid-like collagen comprises broad hypocellular bundles with brightly eosinophilic hyalinization, and myxoid stroma is defined as an amorphous stromal substance with slightly basophilic material. Stroma was regarded as mature when it was composed of fine mature collagen fibers stratified into multiple layers and did not contain keloid-like collagen or myxoid stroma. Intermediate stroma showed keloid-like collagen bundles intermixed with mature stroma, typically having parallel orientation to the mature collagen fi typ. Stroma with myxoid changes was regarded as the immature stroma. The stromal type was classified according to the most immature stromal area observed.[7],[9],[10] CAF phenotypes were divided into two groups according to their morphology on H and E-stained slides. CAFs were regarded as mature when they showed thin, wavy, and small spindle-shaped morphology as in normal fibroblasts, and immature when they showed large, plump spindle-shaped cell with prominent nucleoli. When the proportion of immature fibroblasts was more than 50%, the sample was regarded as having an immature CAF phenotype.[11]


The immunohistochemical staining was performed as previously described.[11] We use 4-μm section that was mounted on coated glass slide for immunostaining. The primary antibodies used in this study were mouse anti-alpha smooth muscle actin-positive (α-SMA) (Abcam; ab7817; 1:500) and mouse anti-desmin (DBioSys; MOB060–05; 1:40) respectively, whereas the secondary antibody was anti-mouse IgG (Novolink™ Polymer Detection System; DS RE 7150-B). The immunostaining of CAFs showed α-SMA positive and desmin negative in the cancer stroma.[12] The FFPE of appendix tissue obtained from appendectomy containing smooth muscle cells was used as the positive control, whereas the negative control was the identical specimen without primary antibody. The expressions of α-SMA and desmin are considered positive if cytoplasma is brown stained by DAB chromogen.

TB is the presence of dedifferentiated single cells or small clusters of up to five cells at the invasive front of colorectal cancer.[6] After selecting an area in which TB was most intensive, they were counted using a ×20 objective lens. Based on the numbers of cells in TB locations, the criteria set by the Japanese classification of CRC were used as in the following: Grade 0; no budding, Grade 1; 1–4 budding, Grade 2; 5–9 budding and Grade 3; 10 or more budding.[4],[7]

All the histopathological evaluations were done by light microscopy with objective lens magnifications of ×4, ×10, and ×20. The histopathological data were categorized based on the following parameters: Invasion depth, lymph node metastasis, LVI, stromal type, CAF type, and TB grade.

Statistical analysis

The associations between several clinicopathologic parameters, such as invasion depth, lymph node metastasis, LVI, with TB grade were evaluated using Fisher's exact test, and the associations of stroma and CAF phenotypes with TB grade were evaluated using the Mann–Whitney nonparametric U-test (IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp). A value of P < 0.05 was considered statistically significant.

   Results Top

Clinicopathological characteristics

The clinicopathological characteristics of 23 colorectal adenocarcinoma specimens are summarized in [Table 1]. The median age was 55 years (range 17–82) and was slightly more common in women: We found 11 (47.8%) male and 12 (52.2%) female patients. Twenty-one specimens showed well-differentiated tumor. The invasion depth, numbers of lymph node metastases, and extent of LVI were also recorded as established prognostic factors of CRCs. Pathology analysis for evaluating and categorizing stromal type, CAF type, and TB grade [Figure 1], [Figure 2], [Figure 3] found 11 (47.8%) samples with immature stroma, 17 (73.9%) with immature CAFs, and 11 (47.8%) with Grade 3 TBs. Using immunohistochemical expression of α-SMA and desmin as CAFs markers, we also confirm that the cells evaluated for morphology are CAFs. As shown in [Figure 4], the immunostaining of α-SMA was positive, while desmin was negative.
Table 1: Clinicopathological characteristics

Click here to view
Figure 1: Stromal phenotypes: (a) stroma with myxoid changes regarded as immature; (b) intermediate stroma, keloid-like collagen fibers were intermixed with mature stroma; (c) mature stroma, fine mature collagen fibers were stratified into multiple layers. Light microscopy, ×10 objective lens; H and E staining

Click here to view
Figure 2: Cancer-associated fibroblast phenotypes: (a) immature cancer-associated fibroblasts appeared as large, plump spindle-shaped cell (black arrow); (b) mature cancer-associated fibroblasts had thin, wavy, and small spindle-shaped morphology (yellow arrow). Light microscopy, ×20 objective lens; H and E staining

Click here to view
Figure 3: Tumor budding grades: (a) low-grade tumor budding showed fewer than five buds (black arrow) in the evaluated area; (b) high-grade tumor budding showed five or more buds (yellow arrow) in the evaluated area. Light microscopy, ×20 objective lens; H and E staining

Click here to view
Figure 4: Immunohistochemistry: (a) positive expression of anti-alpha smooth muscle actin in cytoplasmic of cancer-associated fibroblasts (black arrow); (b) positive control of alpha smooth muscle actin in cytoplasmic of smooth muscle (black arrow); (c) negative control of alpha smooth muscle actin; (d) negative expression of desmin in cancer-associated fibroblasts of colorectal carcinoma stroma; (e) positive control of desmin in cytoplasmic of smooth muscle in muscularis mucosae, wall of blood vessel and muscularis propria; (f) negative control of desmin. Light microscopy, ×10 objective lens

Click here to view

The association of tumor budding grade with invasion depth, lymph node metastasis, and lymphovascular invasion

To determine the association of TB grade with prognostic parameters such as invasion depth, lymph node metastasis, and LVI, we categorized the invasion depth into pT1–T2 and pT3–T4, whereas the grade of TB was classified into low and high grades. [Table 2] shows the significant association (P < 0.01) of TB with the lymph node metastasis and LVI.
Table 2: Association of tumor budding grade with invasion depth, lymph node metastases, and lymphovascular invasion

Click here to view

The association of stromal type and cancer-associated fibroblast phenotype with tumor budding grade

To determine the association of stroma and CAF phenotype with TB grade, stromas were categorized into immature/intermediate and mature types. [Table 3] shows that the CAF phenotype had a significant association with TB grade (P < 0.01); conversely to the association between stromal type and the grade of TB.
Table 3: Association of stromal and cancer-associated fibroblast type with the tumor budding grade

Click here to view

   Discussion Top

Because of the failure of the TNM system to predict the survival of patients with CRCs accurately at the same stages, especially for intermediate-stage CRCs, there is a need to find factors which could better differentiate the prognosis. We found here that the CAF phenotype based on morphology was associated significantly with the TB grade. Studies evaluating the significance of CAF phenotypes in CRCs are still limited. A study on esophageal squamous cell carcinomas (SCCs) revealed that immature CAFs were associated with patient survival.[11] The CAF phenotype can be observed by simple histopathology with H and E staining and needs no ancillary staining. Unfortunately, these observations have not been included in routine histopathology reports. The crosstalk between tumor cells and the normal cells surrounding them might also become an additional prognostic information for predicting tumor behavior.

In our study, the CRC samples showed almost the same proportion between low-and high-grade TB rates. This result differs from that of Satoh et al. in 2014, who found that low-grade TB was found in most (72.7%) of 139 tumors.[4] TB in CRCs is a morphologic manifestation of EMT [5] which is proposed as a critical mechanism for the acquisition of malignant phenotypes by epithelial cells.[13] Statistical analysis of our results revealed that high-grade TB was associated significantly with lymph node metastasis and LVI, which have been established as prognostic factors for CRCs and are commonly used in daily pathology. Another study reported that there is an association between TB and other histopathological factors such as more advanced tumor grade, infiltrating tumor borders, the presence of lymph node metastasis, and LVI in CRCs.[14] TB occurs predominantly in the invasive front of tumors,[13] suggesting that this process is required to support the cancer cells when moving from the main tumor mass and invading the surrounding stroma. All these features correspond to increased tumor cell mobility.[15] (18)The higher the TB grade, the more likely the tumor cells will undergo stromal invasion and metastasis.

According to the histological classification of CAFs, most of our samples demonstrated an immature CAF phenotype and there was a statistically significant association between the CAF phenotype and the grade of TB. The more immature the CAFs, the higher grade of TB. However, data about the significance of CAF phenotypes based on the histologic classification in CRC are still limited although CAF phenotypes can be evaluated in daily practice. Ha et al.[11] found that from a total 116 FFPE tumor samples of esophageal SCCs, 52 cases (44.8%) were classified as having a mature CAF phenotype and 64 (55.2%) as immature phenotype. They also revealed that the incidence of immature CAFs correlated significantly with a complete EMT phenotype. The incidence of immature CAFs compared with mature CAFs also correlated strongly with decreased overall survival and disease-free survival rates.[11]

The significant association of CAF subtype with TB in our results indicates that the CAF phenotype might be a useful predictor for the prognosis of patients with CRCs. It indicates the crosstalk between stroma and cancer cell that plays important roles in cancer progression. However, we could not demonstrate a significant association of CAF phenotype with lymph node metastasis or LVI or any significant association between stromal type and TB grade. Nevertheless, the immature/intermediate stromal types containing keloid-like collagen tended to have higher grades of TB. Evaluation of CAF phenotypes as part of the stromal component and TB seen by histopathology can add to the prognostic information for patients with CRC. Future cohort study is required to obtain the follow-up data for the establishment of the prognostic marker.

   Conclusion Top

The CAF phenotype could be considered as a novel prognostic determinant for patients with CRCs due to its association with TB grade. Future studies are required to examine the secretomes of CAFs that might play important roles in EMT and help further in determining the prognosis for patients with CRCs.

Financial support and sponsorship

This study was supported by a grant from the Ministry of Research.

Conflicts of interest

There are no conflicts of interest.

   References Top

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424.  Back to cited text no. 1
Tárraga López PJ, Albero JS, Rodríguez-Montes JA. Primary and secondary prevention of colorectal cancer. Clin Med Insights Gastroenterol 2014;7:33-46.  Back to cited text no. 2
Takatsuna M, Morohashi S, Yoshizawa T, Hirai H, Haga T, Ota R, et al. Myofibroblast distribution is associated with invasive growth types of colorectal cancer. Oncol Rep 2016;36:3154-60.  Back to cited text no. 3
Satoh K, Nimura S, Aoki M, Hamasaki M, Koga K, Iwasaki H, et al. Tumor budding in colorectal carcinoma assessed by cytokeratin immunostaining and budding areas: Possible involvement of c-met. Cancer Sci 2014;105:1487-95.  Back to cited text no. 4
Lugli A, Kirsch R, Ajioka Y, Bosman F, Cathomas G, Dawson H, et al. Recommendations for reporting tumor budding in colorectal cancer based on the international tumor budding consensus conference (ITBCC) 2016. Mod Pathol 2017;30:1299-311.  Back to cited text no. 5
Koelzer VH, Zlobec I, Lugli A. Tumor budding in colorectal cancer – Ready for diagnostic practice? Hum Pathol 2016;47:4-19.  Back to cited text no. 6
Ueno H, Shinto E, Shimazaki H, Kajiwara Y, Sueyama T, Yamamoto J, et al. Histologic categorization of desmoplastic reaction: Its relevance to the colorectal cancer microenvironment and prognosis. Ann Surg Oncol 2015;22:1504-12.  Back to cited text no. 7
Gascard P, Tlsty TD. Carcinoma-associated fibroblasts: Orchestrating the composition of malignancy. Genes Dev 2016;30:1002-19.  Back to cited text no. 8
Ueno H, Shinto E, Kajiwara Y, Fukazawa S, Shimazaki H, Yamamoto J, et al. Prognostic impact of histological categorisation of epithelial-mesenchymal transition in colorectal cancer. Br J Cancer 2014;111:2082-90.  Back to cited text no. 9
Crispino P, De Toma G, Ciardi A, Bella A, Rivera M, Cavallaro G, et al. Role of desmoplasia in recurrence of stage II colorectal cancer within five years after surgery and therapeutic implication. Cancer Invest 2008;26:419-25.  Back to cited text no. 10
Ha SY, Yeo SY, Xuan YH, Kim SH. The prognostic significance of cancer-associated fibroblasts in esophageal squamous cell carcinoma. PLoS One 2014;9:e99955.  Back to cited text no. 11
Nishishita R, Morohashi S, Seino H, Wu Y, Yoshizawa T, Haga T, et al. Expression of cancer-associated fibroblast markers in advanced colorectal cancer. Oncol Lett 2018;15:6195-202.  Back to cited text no. 12
Zlobec I, Lugli A. Epithelial mesenchymal transition and tumor budding in aggressive colorectal cancer: Tumor budding as oncotarget. Oncotarget 2010;1:651-61.  Back to cited text no. 13
Mitrovic B, Schaeffer DF, Riddell RH, Kirsch R. Tumor budding in colorectal carcinoma: Time to take notice. Mod Pathol 2012;25:1315-25.  Back to cited text no. 14
Grigore AD, Jolly MK, Jia D, Farach-Carson MC, Levine H. Tumor budding: The name is EMT. Partial EMT. J Clin Med 2016;5. pii: E51.  Back to cited text no. 15


  [Figure 1], [Figure 2], [Figure 3], [Figure 4]

  [Table 1], [Table 2], [Table 3]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
    Materials and Me...
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded22    
    Comments [Add]    

Recommend this journal