Elsevier

Human Pathology

Volume 62, April 2017, Pages 91-98
Human Pathology

Original contribution
Analysis of KRAS, NRAS, PIK3CA, and BRAF mutational profile in poorly differentiated clusters of KRAS-mutated colon cancer,☆☆

https://doi.org/10.1016/j.humpath.2016.12.011Get rights and content

Highlights

  • High number of PDCs is associated with nodal metastases and poor prognosis in CRC.

  • PDCs reflect epithelial-mesenchymal transition in CRC and represent the most likely source of cancer metastatization.

  • PDCs may have biomolecular profile different from that of the main tumor tissue.

  • Heterogeneous biomolecular profile in different parts of CRC may be related to different grades of differentiation of the neoplastic cells.

Summary

Recently, a grading system based on the counting of poorly differentiated clusters (PDCs) of neoplastic cells was shown to be a strong predictor of nodal metastases and negative prognosis in colon cancer (CC). In this study, we assessed and compared the mutational status of KRAS, NRAS, and PIK3CA in PDCs and corresponding main tumor tissue of 25 CCs with KRAS mutations. For each tumor, PDC and main tumor tissue were distinctly analyzed by using laser microdissection and mass spectrometry. In 3 CCs, the main tumor tissue had also PIK3CA mutations (C420R: 1; E545K: 1; H1047R: 1), and in 1, it showed NRAS mutation (codon 12). In 20 cases, PDCs had the same biomolecular profile as the main tumor, but in 5, they had different biomolecular profiles. In detail, PDCs had KRAS wild type in 2 cases and additional PIK3CA mutations (E542K: 1; H1047Y: 1; E545Q: 1) in 3. All 3 cases with additional PIK3CA mutations in PDCs had nodal metastases, high pathological TNM stage, and lymphatic invasion. In 1 of 3 cases, additional PIK3CA mutation detected in PDC, but not in the main tumor, was also found in the corresponding nodal metastases. Our findings show for the first time that heterogeneous biomolecular profile previously observed in CC may depend on different histologic aspects of the lesion. Because PDCs may represent the tumor cells with the highest potential to metastatize, their molecular status may be relevant for the prediction of response to targeted therapies.

Introduction

Colorectal cancer (CRC) is one of the most common malignancies and a leading cause of cancer-related death worldwide [1]. At present, the pathological TNM (pTNM) stage assessed in accordance to the International Union Against Cancer TNM [2] and to the American Joint Committee on Cancer [3] represents the most relevant prognostic factor for CRC [4] and determines its postsurgical treatment. However, in some cases, pTNM stage has no correlation with the biological and clinical behavior of CRC [5], [6]. Hence, the identification of additional factors, which may predict the clinical course of CRC with more accuracy and regardless of pTNM stage, has been a main research focus over the years.

Recently, ours and other groups showed that a novel grading system based on the number of clusters composed of at least 5 cancer cells and lacking a gland-like structure—so-called poorly differentiated clusters (PDCs)—represents a significant prognostic factor in patients with CRC, independently of pTNM stage [6], [7], [8], [9], [10], [11], [12], [13]. In addition, the presence of a high number of PDCs was significantly associated with higher metastatic risk of CRC [7], [8], [9], [10], [11], [12]. This association may be explained by the relationship between PDC and epithelial-mesenchymal transition (EMT) [14], [15], a process by which neoplastic cells lose epithelial properties and acquire the mesenchymal cell potential to migrate through the extracellular matrix and to metastatize. Indeed, PDCs show several characteristics of EMT phenotype, such as reversed pattern of MUC1 expression, loss of or aberrant E-cadherin expression, up-regulation of WNT/β-catenin signaling pathway, and matrix metalloproteinase expression [15], [16], [17]. Thus, they have invasive properties similar to those seen in cells forming tumor budding foci [15]. However, because they are composed of at least 5 neoplastic cells [7], PDCs are more easily identified with conventional hematoxylin and eosin stain [18].

In a recent study, we found that CRCs with high counts of PDCs display mutations in KRAS gene with significantly higher frequency than CRCs with absent/low number of PDCs [19]. We speculated that this association may depend on the ability of RAS oncogenic mutations to induce dedifferentiation of CRC cells and EMT process [20], [21]. However, whether KRAS mutations actually exert a role in driving PDC formation is still to be determined. In addition, the molecular signature of PDCs has not been analyzed thus far.

On this premise, aims of the present study were as follows: (a) to investigate the mutational status of genes involved in the RAS/MAPK and PI3K-PTEN-AKT signaling pathways in the PDCs of a series of colonic carcinomas (CCs) showing KRAS mutations and (b) to compare the mutational status observed in PDCs with that revealed in the corresponding main tumor.

Section snippets

Materials and methods

Twenty-five consecutive KRAS-mutated surgically resected CCs with more than 10 PDCs at the invasive front of growth (PDC G3 tumors) were taken from the files of the Unit of Pathological Anatomy of Modena, Italy. All cases were anonymously collected, and all procedures were performed in accordance with the Helsinki Declaration. All relevant issues were discussed with the local ethics committee that established that no further ethical approval was needed to revise histology or to perform

Results

Clinical and pathological features of CCs included in the study are summarized in Table 1.

We observed KRAS mutations in the main component of the CCs at the following sites: at codon 12 in 18 (72%) cases (G12A in 4 cases, G12C in 1, G12D in 11, G12R in 1, and G12S in 1), at codon 13 in 2 (8%) (G13C in 1 case and G13D in 1 case), at codon 61 (Q61H) in 2 (8%), at codon 59 in 1 (4%), at codon 146 (A146T) in 1 (4%), and at codon 117 in 1 (4%). In addition to KRAS mutations, the main tumor component

Discussion

Because of their ability to block downstream intracellular signaling of epidermal growth factor receptor (EGFR), monoclonal antibodies against EGFR are currently used as a therapy for CRC [22]. However, their therapeutic efficacy strictly depends on the integrity of genes involved in the pathways downstream to EGFR, that is, RAS/MAPK and PI3K-PTEN-AKT pathways. Indeed, mutations in KRAS, NRAS, or PIK3CA genes may activate the pathway in the absence of extracellular stimuli [23]. For this

Acknowledgments

We wish to thank Marzia Gozzoli and Rosa Zaramella for their technical support in the molecular analysis.

References (40)

  • V. Barresi et al.

    Stage I colorectal carcinoma: VEGF immunohistochemical expression, microvessel density, and their correlation with clinical outcome

    Virchows Arch

    (2010)
  • V. Barresi et al.

    Immunohistochemical assessment of lymphovascular invasion in stage I colorectal carcinoma: prognostic relevance and correlation with nodal micrometastases

    Am J Surg Pathol

    (2012)
  • V. Barresi et al.

    Colorectal carcinoma grading by quantifying poorly differentiated cell clusters is more reproducible and provides more robust prognostic information than conventional grading

    Virchows Arch

    (2012)
  • H. Ueno et al.

    New criteria for histologic grading of colorectal cancer

    Am J Surg Pathol

    (2012)
  • H. Ueno et al.

    Site-specific tumor grading system in colorectal cancer: multicenter pathologic review of the value of quantifying poorly differentiated clusters

    Am J Surg Pathol

    (2014)
  • H. Ueno et al.

    Novel risk factors for lymph node metastasis in early invasive colorectal cancer: a multi-institution pathology review

    J Gastroenterol

    (2014)
  • J.W. Kim et al.

    Clinicopathologic impacts of poorly differentiated cluster-based grading system in colorectal carcinoma

    J Korean Med Sci

    (2015)
  • H. Ueno et al.

    Prognostic impact of histological categorisation of epithelial-mesenchymal transition in colorectal cancer

    Br J Cancer

    (2014)
  • V. Barresi et al.

    Micropapillary pattern and poorly differentiated clusters represent the same biological phenomenon in colorectal cancer: a proposal for a change in terminology

    Am J Clin Pathol

    (2014)
  • Y. Kajiwara et al.

    Heterogeneity of metalloproteinase expression in colorectal cancer—relation of molecular findings to basic morphology

    Anticancer Res

    (2011)
  • Cited by (19)

    • Development of a multiplex allele-specific qPCR approach for testing PIK3CA mutations in patients with colorectal cancer

      2022, Heliyon
      Citation Excerpt :

      Together with the most frequent activating mutations E542K, E545K, H1047L, and H1047R, several additional mutations were selected for screening based on their frequency in previously published studies and confirmed functional significance [36]. Using the criteria mentioned above, mutations R88Q [37], N345K [37], C420R [38], Q546K [3], M1043I [39], H1047Y [38] (Table 3) were tested in the present study, as they were previously identified in samples from patients with CRC. A mutation E545A [40], characterized as activating, was discarded from the present study, as it was detected in the PIK3CA pseudogene [41].

    • Histopathological variables in liver metastases of patients with stage IV colorectal cancer: potential prognostic relevance of poorly differentiated clusters

      2018, Human Pathology
      Citation Excerpt :

      First, primary CRCs with high number of PDCs have higher probability of having nodal metastases [14] and even nodal micrometastases [31]. Second, when PDCs and main tumor mass have different biomolecular profile, nodal metastases may have the same genetic mutations as those observed in PDCs [32], which suggest that PDCs may represent the source of metastatic disease [32]. Interestingly, we found PDCs in at least 1 LM in most of our patients.

    View all citing articles on Scopus

    Competing interests: We have no conflict of interest to declare.

    ☆☆

    Funding/Support: This study was supported by “Programma di Ricerca Regione Università 2010-2012,” Regione Emilia Romagna, Italy, titled “Use of biomarkers and gene expression profiles to identify cancer patients with different prognosis and sensitivity to molecular targeted agents.”

    1

    These authors contributed equally to this work.

    View full text