Dynamics of RAS Mutations in Liquid Biopsies in Metastatic Colorectal Cancer Patients—Case Series and Literature Review

Liquid biopsies can accurately identify molecular alterations in patients with colorectal cancer with high concordance with tissue analysis and shorter turnaround times. Circulating tumor (ct) DNA analysis can be used for diagnosing and monitoring tumor evolution in patients with metastatic colorectal cancer who are treated with EGFR inhibitors. In this article, we reported three clinical cases to illustrate the relevance of RAS mutations identified in ctDNA samples of patients with wild-type metastatic colorectal cancer who received an EGFR inhibitor plus chemotherapy as first-line treatment. The identification of RAS mutations in these patients is one of the most frequently identified mechanisms of acquired resistance. However, detecting a KRAS mutation via liquid biopsy can be caused by inter-tumor heterogeneity or it can be a false positive due to clonal hematopoiesis. More research is needed to determine whether ctDNA monitoring may help guide therapy options in metastatic colorectal cancer patients. We performed a literature review to assess the technologies that are used for analysis of RAS mutations on ctDNA, the degree of agreement between tissue and plasma and the importance of tissue/plasma discordant cases.


Introduction
Colorectal cancer (CRC) is the third most common and the second most deadly cancer worldwide, representing 10.7% of new cases and 9.5% of cancer-related deaths [1].The treatment of CRC is mainly driven by the stage of the cancer, the patient's clinical condition and certain molecular characteristics of the tumor.In patients with metastatic disease, the preferred therapeutic option is systemic therapy, which may include chemotherapy, targeted agents or immunotherapy.In the era of personalized medicine, it is essential to identify biomarkers that will inform the choice of appropriate therapeutic agents.
This process entails the examination of genomic alterations, including KRAS/NRAS/BRAF mutations, and microsatellite instability (MSI)/mismatch repair genes (MMR).Consequently, patients exhibiting wild-type KRAS, NRAS and BRAFV600E genotypes are likely to benefit from the administration of chemotherapy doublets, including FOLFOX (5-Fluouracil, folinic acid and oxaliplatin), CAPOX (capecitabine and oxaliplatin) and FOLFIRI (5-Fluouracil, folinic acid and irinotecan), in conjunction with EGFR-inhibitors such as Cetuximab or Panitumumab [2].Anti-EGFR monoclonal antibody (moAb) Cetuximab comprises a chimeric immunoglobulin G1 (IgG1), which, upon binding to the EGFR receptor, initiates the internalization and degradation of the receptor, thus disrupting the downstream pathway.Cetuximab is also capable of eliciting immune functions such as antibody-dependent cell-mediated cytotoxicity (ADCC), which serves as an anti-tumor effect [3].Nevertheless, Panitumumab, a humanized IgG2 monoclonal antibody with a similar mechanism of action, is unable to elicit an ADCC effect, but in return it makes it less likely to determine secondary reactions [4].
The detection of molecular alterations has enabled personalized treatment in oncological diseases.Mutations in RAS and BRAF genes are widely recognized as the primary drivers of resistance to anti-EGFR agents in metastatic colorectal cancer (mCRC).Testing for these mutations is essential as they serve as predictive biomarkers for both primary and acquired resistance [5].Primary resistance refers to the presence of activating mutations in these genes prior to treatment initiation, while the selection of pre-existing mutant clones or de novo acquisition of mutations under the influence of anti-EGFR treatment is known as acquired resistance [6].The identification of molecular mechanisms behind acquired resistance to cetuximab or panitumumab in mCRC is crucial for predicting the failure of anti-EGFR antibodies and developing new treatment approaches.
Cancers are defined by the simultaneous presence of various genetic changes, resulting in the occurrence of spatial and temporal heterogeneity [6,7].While tissue biopsy is the gold standard and preferred method for cancer diagnosis, it may not fully capture the spatial and temporal heterogeneity of the tumor.Tissue biopsies are invasive procedures, associated with risks of complications, low patient compliance and substantial procedural expenses [6].
When compared to repeated tumor tissue biopsies, the use of circulating tumor (ct) DNA for genotyping the tumor (liquid biopsy) is less invasive and does not require a lesion accessible to biopsy.Furthermore, liquid biopsy is able to capture the tumor heterogeneity that appears during treatment before radiological progression [8].Thus, liquid biopsy could be a valuable method for assessing the presence and heterogeneity of the tumor, as well as monitoring cancer treatment.
Studies have demonstrated that ctDNA analysis can accurately identify molecular alterations in patients with CRC with high concordance with tissue analysis and shorter turnaround times [9][10][11][12][13][14].Recent studies have demonstrated the utility of ctDNA analysis for diagnosing and monitoring tumor evolution in patients with mCRC who are treated with EGFR inhibitors.In this patient population, RAS/BRAF-acquired mutations were detected at the time of disease progression [15][16][17].
The case series reported in this article aim to illustrate the clinical relevance of RAS mutations identified in ctDNA samples of patients with mCRC with a baseline RAS wildtype tumor tissue and who received an anti-EGFR monoclonal antibody plus chemotherapy as first-line treatment.Furthermore, we performed a literature review to assess the technologies that are used for analysis of RAS mutations on ctDNA, the degree of agreement between tissue and plasma and the importance of tissue/plasma discordant cases.

Case Series
The patients reported in this article were enrolled in an ongoing prospective study conducted at our institution (Fundeni Clinical Institute) that aims to assess the mutational status of KRAS, NRAS and BRAF on liquid biopsy at predefined intervals using the Idylla© Biocartis (Mechelen, Belgium)system in patients with mCRC (Figure 1).Mutational analysis was performed using the molecular biology equipment real time ( The performance of the rtPCR Idylla© Biocartis platform as a method for determining these mutations has been highlighted by numerous studies [14,18].opsy)-NRAS (18 mutations)-Exon 2 (codon 12, 13), Exon 3 (codon 59, 61), Ex 117, 146); BRAF (5 mutations)-Exon 15, codon 600 (BRAF V600E, BRAF V V600K, BRAF V600 R) The performance of the rtPCR Idylla© Biocartis platform as a method fo ing these mutations has been highlighted by numerous studies [14,18].

Literature Search Strategy
A literature search was conducted in PubMed and Web of Science datab ruary 2024, using the following key terms: colorectal cancer, colon cancer, m orectal cancer/mCRC, RAS, KRAS, NRAS, liquid biopsy, circulating tumor DN DNA.Relevant medical subject headings terms were used.The titles and ab articles were assessed, and irrelevant studies were eliminated.Relevant artic tained in full text and evaluated.Hand searches of the reference lists were als to identify any relevant articles that were missed with the search strategy.

Case 1: A 60-Year-Old Male Diagnosed with Sigmoid Colon Cancer with Peritoneal Carcinomatosis
The initial pathological evaluation of the biopsy obtained from the pri revealed a poorly differentiated adenocarcinoma.RAS/BRAF testing on tissu time PCR showed a wild-type tumor.The patient received first-line Cetuxima FOX (fluorouracil, leucovorin, oxaliplatin), with stable disease at two-mont month follow-up.At the time of each imaging evaluation, liquid biopsy w formed using Idylla© Biocartis, a fully automated real-time-PCR-based mol nostic system and it confirmed the KRAS, NRAS and BRAF wild-type status The CT scan performed three months later showed progressive disease ous liquid biopsy detected a KRAS mutation (G12V) at the time of progress tient then started treatment with Bevacizumab and FOLFIRI and had stable d first follow-up.After six months, the patient's performance status deteriora was referred to palliative care.

Literature Search Strategy
A literature search was conducted in PubMed and Web of Science databases in February 2024, using the following key terms: colorectal cancer, colon cancer, metastatic colorectal cancer/mCRC, RAS, KRAS, NRAS, liquid biopsy, circulating tumor DNA, cell-free DNA.Relevant medical subject headings terms were used.The titles and abstracts of all articles were assessed, and irrelevant studies were eliminated.Relevant articles were obtained in full text and evaluated.Hand searches of the reference lists were also performed to identify any relevant articles that were missed with the search strategy.

Case 1: A 60-Year-Old Male Diagnosed with Sigmoid Colon Cancer with Peritoneal Carcinomatosis
The initial pathological evaluation of the biopsy obtained from the primary tumor revealed a poorly differentiated adenocarcinoma.RAS/BRAF testing on tissue using real-time PCR showed a wild-type tumor.The patient received first-line Cetuximab plus FOLFOX (fluorouracil, leucovorin, oxaliplatin), with stable disease at two-month and four-month follow-up.At the time of each imaging evaluation, liquid biopsy was also performed using Idylla© Biocartis, a fully automated real-time-PCR-based molecular diagnostic system and it confirmed the KRAS, NRAS and BRAF wild-type status (Table 1).
The CT scan performed three months later showed progressive disease.Simultaneous liquid biopsy detected a KRAS mutation (G12V) at the time of progression.The patient then started treatment with Bevacizumab and FOLFIRI and had stable disease at the first follow-up.After six months, the patient's performance status deteriorated, and he was referred to palliative care.At time of the initial diagnosis, the pathological evaluation of the biopsy specimen from the primary tumor showed a moderately differentiated adenocarcinoma.Genomic testing revealed a KRAS, NRAS, BRAF wild type and a dMMR tumor.RAS/BRAF testing on tissue was performed with real-time PCR, while the MMR status was evaluated by immunohistochemistry. Concomitantly, ctDNA mutational analysis was performed using the Idylla© Biocartis.Surprisingly, the liquid biopsy analysis at baseline detected a KRAS G13D mutation (Table 1).
Given the RAS/BRAF wt status on tissue analysis (which represents the standard of care), the patient received first-line treatment with Cetuximab and FOLFOX.A control MRI was performed three months after systemic treatment initiation and showed a partial response to treatment.Simultaneously, liquid biopsy was repeated, and it revealed this time a KRAS G12D mutation (Table 1).The patient continued the same treatment and the imaging evaluation three months later showed stable disease, while the liquid biopsy detected the same KRAS G12D mutation.
The patient received maintenance treatment with cetuximab and deGramont (fluorouracil, leucovorin).Three months later (9 months from the initial diagnosis), the clinical status of the patients deteriorated rapidly, and the CT scan showed progressive disease.At the time of the imaging evaluation, ctDNA mutational analysis was also repeated and a KRAS G12D mutation was detected (Table 1).The patient was referred to palliative care and died within one month.

Case 3: A 68-Year-Old Female Diagnosed with Sigmoid Colon Cancer with Synchronous Lung and Liver Metastasis
At time of diagnosis, the pathological evaluation of a biopsy specimen from the primary tumor showed a moderately differentiated adenocarcinoma.KRAS, NRAS and BRAF testing on tissue was performed with real-time PCR and revealed a wild-type tumor.At the same time, ctDNA mutational analysis with Idylla© Biocartis confirmed the wildtype status.The patient started treatment with cetuximab and FOLFOX and had partial response at three-month follow up.However, liquid biopsy at the time identified a KRAS mutation (K117N).Patient continued the same treatment and had stable disease at sixmonth follow up.RAS/BRAF testing using Idylla© Biocartis was repeated at the time of the imaging evaluation and revealed this time a wild-type status.The patient continued maintenance treatment with cetuximab and 5-FU/LV (fluorouracil, leucovorin).Imaging evaluation at nine-month and twelve-months follow-up showed stable disease and the liquid biopsy did not detect any KRAS/NRAS/BRAF mutation.The patient is currently undergoing the same treatment and imagining evaluations and ctDNA mutational analysis will be continued.

Literature Review
We performed a literature review to assess the degree of agreement between tissue and plasma RAS mutational status in patients with mCRC and the importance of tissue/plasma discordant cases.The main focus was the identification of a KRAS mutation in the plasma of patients diagnosed initially with wild-type mCRC.
Circulating tumor DNA (ctDNA) levels are found to correlate with a number of biological and clinicopathological characteristics, including tumor burden, stage, histotype, apoptotic rate, proximity to blood vessels, and metastatic potential [19][20][21].A considerable proportion of patients with mCRC display detectable ctDNA in plasma, with mutation rates ranging from 1.9% to 27% [20].Consequently, the non-invasive identification of emerging KRAS mutations in cell-free DNA (cfDNA) from peripheral blood can facilitate the detection of resistance to anti-EGFR therapy [22].
High concordance between liquid biopsy and tissue biopsy for tumor molecular profiling has been previously demonstrated [23].A prospective study compared KRAS mutation results obtained from plasma cfDNA testing with those obtained from tumor tissue biopsy and showed that plasma cfDNA analysis exhibited 98% specificity and 92% sensitivity, with a concordance rate of 96% [24].Numerous studies demonstrated a good concordance between tissue and cfDNA testing of RAS mutations in patients with mCRC [11,[25][26][27][28]. Several studies focused on patients with a RAS-positive liquid biopsy but a negative tissue RAS test.A comprehensive presentation of these articles can be found in Table 2.A RAS mutation was identified by liquid biopsy in patients with wild-type mCRC on tissue analysis either at baseline or during treatment with EGFR inhibitors.The rate of RAS positive cases on liquid biopsy at baseline or following treatment with an anti-EGFR in patients with wild-type mCRC differs across studies, most likely due to the limited number of patients and the different technologies that were used for testing.
The identification of mutations in ctDNA from liquid biopsies requires sensitive techniques.Droplet digital PCR (ddPCR) is a highly sensitive, specific and precise technique that facilitates the identification and quantification of specific DNA alterations in various clinical specimens [37].Recent studies have shown the efficacy of ddPCR in detecting RAS and BRAF mutations in metastatic colorectal cancer (mCRC).The Poseidon study compared liquid biopsy using ddPCR with standard tissue-based molecular testing for RAS/BRAF genotyping in mCRC patients.The liquid biopsy approach showed faster results (7 days vs. 22 days) and high concordance with standard tissue analysis (83%).This supports the use of liquid biopsy and ddPCR in routine mCRC care, especially for rapid decision-making on first-line therapy [38].Iris van 't Erve et al. [39] examined the use of liquid biopsy RAS/BRAF ctDNA analyses in mCRC patients to determine eligibility for anti-EGFR therapy.The results showed a 93% concordance between tissue DNA and liquid biopsy ctDNA mutations, with ddPCR being a part of the diagnostic strategy [39].When compared to next-generation sequencing (NGS) technology, Demuth et al. [40] concluded in a study measuring KRAS mutations in ctDNA of CRC patients using ddPCR and NGS that both methods have high concordance to tumor genotype (79% and 89%).This study also showed that, in cases with sparse material from mCRC patients, smaller plasma volumes may often be sufficient for KRAS mutation detection by ddPCR [40].
Extended RAS and BRAF mutation analysis using NGS was also explored, establishing the importance of the method in analyzing these mutations.In contrast with PCR-based methods, which only detect known mutations in certain genes, NGS has the benefit of analyzing multiple types of genomic alterations simultaneously.It does not only identify point mutations, but also detects insertions, deletions, gene amplifications and microsatellite instability [41].In ctDNA sequencing, the sensitivity of detecting mutant allele frequency (MAF) or variant allele frequency (VAF) is pivotal for evaluating the effectiveness of an NGS-based ctDNA profiling assay [42,43].MAF or VAF measures the proportion of DNA molecules carrying a mutation compared to the total number of molecules with the same allele.These metrics indicate the quantity of ctDNA in relation to cell-free DNAs containing tumor-specific mutant alleles.Thus, a lower detectable MAF signifies greater sensitivity of an NGS assay for ctDNA analysis.
The OncoBEAM platform aims to address the technical challenge of detecting rare DNA molecules by implementing improvements to the PCR technique through a process known as BEAMing [26].DNA amplification is used to facilitate measurement and enhance the quantity of DNA species.BEAMing enables the detection of uncommon DNA sequences (e.g., those carrying a RAS mutation) with high resolution by dividing the PCR procedure into numerous individual reactions [26].Bando et al. tested the concordance of the RAS mutational status between plasma ctDNA and tumor tissue in Asian patients diagnosed with mCRC using the OncoBEAM platform.The concordance rate between plasma-and tissue-BEAMing was 86.4% [25].Similar previous studies showed a concordance of 89.7-93.3%, which indicates that the detection of RAS mutations in the blood through liquid biopsy and BEAMing technology may be a useful replacement to tumor testing [26,28,44].
The Idylla system (Biocartis, Mechelen, Belgium) is a fully automated, real-time PCR-based molecular diagnostics system that can rapidly and without preanalytical DNA extraction identify oncogenic mutations, such as in the KRAS gene, in tissue and plasma samples [45].Vitiello et al. investigated the use of liquid biopsy for detecting RAS/BRAF mutations in patients with mCRC using the Idylla™ Biocartis platform to evaluate KRAS, NRAS and BRAF mutations in ctDNA.The overall agreement between liquid biopsy results and standard tissue-based NGS analyses was 81.94% [14].The main characteristics of each method are presented in Table 3.

Discussion
In this case series, we summarized the clinical courses of three RAS wild-type mCRC patients, who were retested for RAS/BRAF mutations during standard-of-care regimens using liquid biopsy.Repeated biopsies to confirm RAS status are not currently used in routine clinical practice, and thus, changes in RAS mutational status during treatment are not completely understood so far.In this case series, all three patients were initially diagnosed as RAS wild-type mCRC, but a KRAS mutation was detected by liquid biopsy.
Genomic instability plays a critical role in carcinogenesis and cancer resistance to specific treatments, and leads to intra-and intertumoral heterogeneity [46].Clonal diversity confers a competitive edge to cancer cells by increasing the likelihood of preexisting resistant clones, thereby promoting acquired resistance.Chemotherapies, particularly targeted therapies, often yield partial clinical responses, creating a selective pressure that can induce redistribution among clonal populations.
Due to tumor heterogeneity, clonal redistribution is difficult to capture in a tissue sample that is fixed in time.Moreover, it is not feasible to perform multiple tissue biopsies to detect such clonal redistribution in time.These limitations can be overcome by liquid biopsies performed on serially collected blood samples from patients.Moreover, liquid biopsies represent a better method to obtain a comprehensive molecular profile of the primary and metastatic tumors.On the other hand, the limitations of liquid biopsy include mainly the reliance of results on tumor volume, and the lack of a standard technique for analysis [47].
There are several possible explanations for the detection of a KRAS mutation by liquid biopsy in the patients reported in this article.Besides primary resistance mechanisms, the efficacy of anti-EGFR agents in mCRC is limited by the development of molecular alterations producing acquired resistance [7,16,48,49].For instance, it has been demonstrated that a significant percentage of patients with KRAS/NRAS wild-type CRC undergoing treatment with EGFR inhibitors will develop RAS mutations at the time of progression [4].CRC may contain resistant mutant clones before treatment, and the proportion of these resistant clones increases under treatment [50].In case 1, we reported a patient with wildtype mCRC who became RAS-mutant at the time of progression.This phenomenon has been confirmed in several reports, although the rate of patients who become RAS-mutant at progression significantly differs among the various studies [10,15,22,30,[51][52][53][54].
In case 2, a KRAS mutation was identified by liquid biopsy although the patient had a partial response to treatment with cetuximab and chemotherapy.There are data suggesting that resistant mutations detected at very low variant allele frequency (VAF) represent small subclones of the total tumor burden and do not predict lack of response to anti-EGFR treatment [15].However, six months after the detection of a KRAS mutation, disease progression was documented by radiological assessment in this patient.Studies have shown that the appearance of resistant KRAS mutant clones could be detected for up to 10 months before radiographic confirmation of disease progression [17].
Tumor heterogeneity may also be responsible for the fact that KRAS mutations were found in liquid biopsies but were missed in tumor tissue.In patients with wild-type RAS primary tumors, 10% of the metastatic lesions have RAS mutations [55].However, current data on the concordance of KRAS mutation status between primary tumors and corresponding metastases are conflicting.Numerous studies investigating inter-tumor heterogeneity (between primary tumors and metastases in the same patient) have demonstrated mutational discordance in up to 30% of cases [24,56,57].The effectiveness of using KRAS mutation status from the primary tumor alone to predict response to anti-EGFR agents remains a subject of ongoing debate.Thus, additional research is needed to gain a deeper understanding of the interplay between the primary tumor and its distant metastases.
Another possibility is that the KRAS mutations detected by liquid biopsy are false positive alterations due to clonal hematopoiesis, and this may be the case for the third case we reported.Previous studies have shown that genes such as TP53 and KRAS can be mutated in white blood cells [58,59].KRAS mutations detected in cfDNA were eventually classified as clonal hematopoiesis mutations, and not tumor-derived mutations [59].Clonal hematopoiesis (CH) arises spontaneously through the accumulation of somatic mutations and the clonal expansion of hematopoietic stem cells during normal aging.The detection of these CH-associated mutations has been characterized as a source of biological stamp in plasma liquid biopsy analyses [58].Misclassification of cfDNA mutations could lead to ineffective therapeutic strategies.Moreover, interference from CH may contribute to false-positive assessments of cfDNA biomarkers, potentially influencing inappropriate treatment decisions.One method to overcome this challenge could be the simultaneous genotyping of cfDNA and white blood cells DNA [60].However, the increased expenses linked with paired sequencing could pose a challenge to the widespread introduction of liquid biopsy in clinical settings.Enhanced understanding of the molecular features of clonal hematopoiesis coupled with machine learning could represent the optimal solution for paired sequencing with white blood cells [61].

Conclusions
Based on available evidence, liquid biopsy represents a novel technique for managing patients with mCRC undergoing treatment with EGFR inhibitors.Several studies have reported a dynamic evolution of RAS/BRAF mutations in mCRC patients treated with first-line EGFR inhibitors and chemotherapy.The identification of RAS mutations in these patients is one of the most frequently identified mechanisms of acquired resistance.However, detecting a KRAS mutation via liquid biopsy can be a false positive due to clonal hematopoiesis.More research is needed to determine whether ctDNA monitoring may help guide therapy options in mCRC patients.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/jpm14070750/s1,Liquid biopsy analysis using Idylla fully automated, real-time PCR based molecular testing system.

Table 2 .
Published studies of dynamic molecular monitoring by liquid biopsy (ctDNA) in patients with RAS wild-type metastatic colorectal cancer.

Table 3 .
Overview of technologies used for detection of KRAS mutation in ctDNA.