Radiol Oncol 2019; 53(1): 85-95. doi: 10.2478/raon-2019-0013 85 research article Molecular biomarkers and histological parameters impact on survival and response to first- line systemic therapy of metastatic colorectal cancer patients Martina Rebersek, Tanja Mesti, Marko Boc, Janja Ocvirk Department of Medical Oncology, Institute of Oncology Ljubljana, Ljubljana, Slovenia Radiol Oncol 2019; 53(1): 85-95. Received 30 August 2018 Accepted 20 September 2018 Correspondence to: Assoc. Prof. Janja Ocvirk, M.D., Ph.D., Department of Medical Oncology, Institute of Oncology Ljubljana, Zaloska 2, SI- 1000 Ljubljana, Slovenia. Phone: +386 1 5879 285: E-mail: jocvirk@onko-i.si Disclosure: No potential conflicts of interest were disclosed. Background. Histological parameters of primary tumour and nodal metastases are prognostic factors for survival of operable colorectal (CRC) patients, but not predictive for response rate of systemic therapy. KRAS mutations in co- dons 12 and 13 were first recognized as a predictive factor for resistance to anti-EGFR monoclonal antibodies. Not all patients with wild-type KRAS (wtKRAS) respond to anti-EGFR antibody treatment. Additional mechanisms of resistance may activate mutations of the other main EGFR effectors pathway, such as other mutations in RAS gene, mutations in P13K and PTEN expression. Patients and methods. In the prospective study prognostic and predictive impact of histological parameters of primary tumour, KRAS and BRAF mutations on overall survival (OS) and objective response (OR) rate of metastatic CRC (mCRC) patients treated with 1st line systemic therapy were analysed. We additionally retrospectively analysed other mutations in RAS genes and their impact on survival and time to progression. Results. From November 2010 to December 2012, we enrolled 154 patients in the study, 95 men and 59 women. Mutations in KRAS gene and V600E BRAF gene were found in 42% and in 3% of patients, respectively. Median OS of the patients with T1, T2 and T3 tumour was 65.4 months (95% CI, 55.7–75.6) while in patients with T4 tumour, lymphangi- osis, vascular and perineural invasion it has not been reached yet. Median OS of the patients with G1, G2 and G3 of tumour differentiation was 65.6 (95% CI, 53.7–77.5) and 25.3 months (95% CI, 16.6–34.1), respectively. Median OS of the patients with stage N0, N1 and N2 was 65.6 (95% CI, 56.4–74.8) and 58.0 months (95% CI, 21.9–94.2), respectively. Median OS of wtKRAS and mutated KRAS patients was 56.5 (95% CI, 48.2–64.9) and 58 months (95% CI, 52.6–63.4), respectively. Median OS of mutated codon 12 and codon 13 patients was 57 (95% CI, 50.9–64.4) and 44 months (95% CI, 40.1–48.4), respectively. Median OS of wtBRAF and of mutated BRAF patients was 59.2 (95% CI, 52.5–65.9) and 27.6 months (95% CI, 12.6–42.5), respectively. wtKRAS significantly affected the response to the first systemic therapy (p = 0.028), while other parameters did not affected it, p= 0.07. In 14 patients (17%), additional mutations in NRAS gene, codon 61 and codon 146 were found. Median OS of wtNRAS, codon 61 and 146 patients was 67.1 months (50.3–67.6) while median OS of mutated NRAS patients has not been reached yet (p = 0.072). Median time to progression of wtNRAS, codon 61 and 146 patients was 11.7 months (10.4–14.5) while median time to progression of mutated NRAS patients was 7.9 months (6.1–11.0), (p = 0.025). Conclusions. Mutated BRAF, N2 and G3 of primary tumour were poor prognostic factors for OS in mCRC patients. wtKRAS significantly affected the response to the first line systemic therapy. Histological parameters included in the analysis and mutated BRAF did not affect significantly the efficacy of 1st line systemic therapy in mCRC patients. Key words: metastatic colorectal cancer; chemotherapy; targeted therapy; histological parameters; biomarkers Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers86 Introduction Colorectal cancer (CRC) is one of the most com- mon cancer and one of the leading causes of cancer death in the world. It is one of the most common cancers in Slovenia and, according to the Cancer Registry of Slovenia, 1353 new patients were diag- nosed with CRC in 2015.1 Metastatic disease is still incurable, with 5% five-year overall survival (OS) without treatment. With the introduction of new chemotherapy, using oxaliplatin and irinotecan in the current management of metastatic disease, in combination with biologicals, targeting epidermal growth factor- mediated growth regulatory path- way and the vascular endothelial growth factor- mediated angiogenesis pathway we can prolong the progression-free survival (PFS) and OS of these patients.2-10 In selected patients with appro- priate combination of therapy and surgery we can achieve approximately a 50% five-year OS. The development of CRC is a multistep process, which accumulates different gene mutations, chro- mosomal abnormalities and epigenetic changes.11 The mutations within KRAS proto-oncogene, pre- dominately within codons 12 and 13, activate RAS/ RAF signalling and are thought to occur early in carcinogenesis of CRC. The KRAS status is the first molecular marker to predict the response to anti-EGFR monoclonal antibodies cetuximab and panitumumab in metastatic CRC (mCRC) patients, and it needs to be determined before deciding in favour of treatment with anti-EGFR antibodies. As the KRAS mutations occur early in CRC for- mation, there is a high concordance between the KRAS mutations of primary tumour and metasta- ses, which was confirmed in previous studies.11 In the retrospective study, de Roock et al. raised the possibility that the patients with the KRAS muta- tion in codon 13 might have benefited from anti- EGFR antibodies treatment.12,13 The mutations in KRAS gene were found in approximately 30 to 40% of mCRC patients, reported in previous literature, but, only 40 to 60% of these patients with wtKRAS will respond to anti-EGFR antibodies treatment. Therefore, other molecular markers downstream of EGFR in the RAS/RAF/MAPK pathway and oth- er effector pathways were found to be involved to predict the response to specific systemic therapy. The BRAF gene encodes a serine/threonine protein kinase of the RAS/RAF/MEK/ERK kinase pathway and it is also involved in CRC carcinogen- esis.11,14,15 The most common mutation of the BRAF gene is V600E which is found in approximately 5 to 9% of mCRC.15,16 The same was reported in our previous study carried on Slovenian patients with CRC where the BRAF V600E mutation was found in 5.1% of patients.16 Previous retrospective studies suggested that mutated BRAF was a marker of re- sistance to anti-EGFR therapy and that the patients with mutated BRAF had significantly shorter PFS and OS than the patients with wtBRAF tumours.14,15 The mutations in the KRAS and BRAF genes have been reported to be mutually exclusive. In the ret- rospective analysis by Fariña- Sarasqueta et al., it was also shown that the BRAF V600E mutation was an independent prognostic factor for the OS of patients with CRC in stages II and III, while the KRAS mutations did not have any effect on the OS of these patients.17 They concluded that the prog- nostic role of the KRAS mutations in an adjuvant setting had to be determined. In published clinical studies the BRAF V600E mutation in mCLC is con- ferred to a poor prognosis regardless of treatment, but these patients may have some benefit from the treatment with cetuximab in combination with chemotherapy as the first-line therapy, except for the patients in whom the disease has progressed after the first-line therapy.15,17 The status of muta- tions in the RAS gene is a new molecular predictive factor for response to treatment with EGFR inhibi- tors in mCRC. These mutations in the RAS gene, in the codons 12, 13, 61 and 146, according to the liter- ature data, are about 15%, and they are determined from autumn 2013 at our Institute of Oncology.3,4,9 The aim of this prospective study was to analyse overall response rate (ORR), time to progression (TTP) and OS of the patients with mCRC treated with first-line systemic therapy in respect of histo- logical parameters of primary tumour KRAS and BRAF status. We additionally retrospectively ana- lysed other mutations in RAS genes and their im- pact on OS and TTP. Patients and methods Patients and treatment In the study, 154 patients with histologically con- firmed mCRC, primarily metastatic or progressed during or after adjuvant therapy were prospec- tively analysed. They were treated according to the national, ESMO and NCCN guidelines, including performance status of patients and comorbidity. They were treated with chemotherapy, including fluoropirimidins, capecitabine or 5-fluorouracil (5- FU), oxaliplatin or irinotecan in combination with biologicals, bevacizumab or cetuximab in respect of previously determined KRAS status. The treat- Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers 87 ment was continued according to the RECIST cri- teria, until the planned operation or until the pro- gression of disease or toxicity occurred. All relevant data from medical files were collected and entered into the data base. Baseline data was analysed with regard to age, sex, primary site (colon and rectum), number and location of metastases. Efficacy was evaluated according to the response evaluation criteria in solid tumours (RECIST, version 1.1) by using computed tomog- raphy (CT) scans, magnetic resonance scans, ab- dominal ultrasound, chest X-ray, bone scans, clini- cal examination and laboratory tests.17 Toxicity was assessed according to the National Cancer Institute common toxicity criteria for adverse events (NCI- CTCAE), version 4.03. The study was approved by the Institutional Review Board Committee and was carried out according to the Declaration of Helsinki. Hystology From the histological findings, data on the tumour size, the radical nature of the primary tumour op- eration, the presence of vascular and perinevral in- vasion, carcinogenic lymphangiosis, the degree of differentiation, the presence of outbreaks of malig- nant cells and the number and severity of regional lymph nodes were collected. Biomarker analysis DNA for molecular analysis was extracted from formalin-fixed, paraffin-embedded tumour tissue of primary tumours or metastases with at least 70% of tumour cells. TheraScreen KRAS Mutation Kit® (Roche Applied Science, Mannheim, D) was used to determine seven most common mutations in codons 12 and 13 of the KRAS gene. The V600E mutation in BRAF was detected by end-point gen- otyping using the TaqMan MGB probes (Applied Biosystems, Warrington, UK) as described previ- ously.16 The mutation V600E in BRAF in positive tumour samples was confirmed by direct sequenc- ing after amplification of the exon 15 of the BRAF gene.16 Retrospectively additional mutations in NRAS gene, in KRAS codon 61 and 146 were de- termined with the same method as mutations in KRAS codon 12 and codon 13. Statistical methods The primary end-points of the analysis were over- all response rate (ORR), based on RECIST criteria, OS and time to progression (TTP) according to the KRAS and BRAF status. The χ2-test was used to compare ORR, OS and TTP between groups, with 95% confidence inter- vals (CI) calculated for the medians. OS and TTP were estimated by using Kaplan-Meier Estimates and compared using the log-rank test. TTP was measured in all patients from the beginning of the first-line systemic chemotherapy to the first evi- dence of progression. The duration of OS was cal- culated from the beginning of systemic treatment until the date of death. P value < 0.05 was consid- ered statistically significant. Statistical data were obtained using the SPSS software package PASW statistics 18.0. The relationship between treatment efficacy and mutations in KRAS and BRAF gene and his- tological characteristics of the tumour (tumour size, presence of vascular and perineural invasion, lymphangiosis, degree of differentiation, number of affected regional lymph nodes) were evaluated by the method of the nominal logistic regression. With the Cox proportional hazards model, we eval- uated the impact of histological factors, the num- ber of affected nodes, and mutations in KRAS and BRAF gene on OS or TTP. Survival was calculated by Kaplan-Meier method to compare groups of pa- tients, and we used the log-rank test. We marked statistically significant differences at p < 0.05. The same statistical methods were used for OS and TTP in the case of retrospectively determined addition- al mutations in RAS gene. Results Patient’s characteristics In total, 154 patients with mCRC who received first-line therapy between November 2010 and December 2012 were included in this prospective analysis. The cut-off date for the present analysis was April 2011. All patients were treated at the Institute of Oncology Ljubljana, all were Caucasian. The median age was 62 years (range 27–86) and the majority of the patients were males (61%). Most of the patients had mCRC (71.4%). One hundred and four patients had primary metastatic disease (59.1%). The most common sites of metastases were liver and lung. The most common therapies the patients received were irinotecan, capecit- abine with bevacizumab (29.5%) and oxaliplatin, capecitabine with cetuximab (22.1%). Twenty-four patients (13.6%) were treated only with chemother- apy, capecitabine in monotherapy, or with fluoro- Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers88 pirimidines in combination with oxaliplatin or iri- notecan. Patients’ baseline characteristics and are shown in Table 1, disease characteristics are shown in Table 2. Median follow-up was 15 months. Of the en- rolled patients, 71% had WHO performance status 1, and 73% of them the CRC primary metastatic disease was confirmed in 58% of patients. Median time to first progression after primary treatment of operable disease was 20 months. Because of a small number of R1 resections of the primary tumour (in 4 patients), perforations of the primary tumour during surgery (in 5 patients) and extramural tu- mour deposits (in 4 patients), these three param- eters were excluded from the analysis. In 42% of patients, the mutations in the KRAS gene were found, most frequently in codon 12 (73%), whereas the V600E mutation in the BRAF gene was found in 3%. The mutations of the KRAS or BRAF gene were detected in total in 76 patients (43.4%). Medium OS of the patients with T1, T2 and T3 primary tumour was 65.4 months (55.7–75.6), while in the patients with T4 primary tumour, it has not been reached yet (p = 0.08). Median OS of the pa- tients without vascular invasion was 65.6 months TABLE 1. Patient’s baseline characteristics Patient’s characteristics Number (%) Medium age (years) 62 Gender female male 59 (38) 95 (62) WHO performance status 0 1 2 36 (23) 109 (71) 9 (6) Tumour location colon rectum 112 (73) 42 (27) Primary metastatic 89 (58) Liver metastases 68 (44) TABLE 2. Disease characteristics Disease characteristics Number (%) pT4 of primary tumour 35 (23) Affected regional lymph nodes (N) N0 (no affected regional lymph nodes) N1 (1 to 3 affected regional lymph nodes) N2 (more than 3 affected regional lymph nodes) 34 (22) 61 (40) 59 (38) Vascular invasion 22 (14) Perineal invasion 25 (16) Lymphangiosis 27 (17) Grade of differentiation G1 (well) G2 (medium) G3 (poorly) 10 (6) 131 (85) 13 (8) KRAS gen non-mutated (wild-type) mutated 89 (58) 65 (42) KRAS mutation codon 12 codon 13 48 (73) 17 (27) BRAF gen non-mutated (wild-type) mutated 150 (97) 4 (3) FIGURE 1. Overall survival of patients according to regional lymph node status (p = 0.000). FIGURE 2. Overall survival of patients according to tumour grade of differention G1 and G2 versus G3. Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers 89 (53.5–77.7), and in the patients with vascular in- vasion, it has not been reached yet. Median OS of the patients without perinevral invasion was 65.6 months (47.9–83.3), and in the patients with peri- nevral invasion, it has not been reached yet. Median OS of the patients without lymphangiosis was 65.6 months (53.5–77.6), and in the patients with lym- phangiosis, it has not been reached yet. Median OS of the patients with grade 1 and 2 of tumour differentiation was 65.6 months (53.7–77.5), and of the patients with grade 3 of tumour differentiation, it was 25.3 months (16.6–34.1) (p = 0069). Median OS of the patients without affected regional lymph nodes or less than 3 of them (N0 and N1) was 65.6 months (56.4–74.8), and of the patients with more than 3 affected regional lymph nodes (N2), it was 58.0 months (21.9–94.2); the difference was statisti- cally significant (p = 0.000). OS of the patients ac- cording to regional lymph nodes and according to the grade of tumour differentiation are shown in Figure 1 and 2, respectively. Median time to first progression after primary treatment of operable disease in the patients with the wtKRAS gene was 20 months (14.7–26.1), and in the patients with the mutated KRAS gene, it was 21 months (17.1–25.4) (p = 0.88). In the patients with the mutations in codon 12, it was 21 months (14.3–28.3), and in the patients with the mutations in codon 13, it was 16 months (11.5–20.7) (p = 0.53). Median time to first progression after primary treatment of operable disease in the patients with the wtBRAF gene was 20 months (16.7–24.1), and in the patients with the mutated BRAF gene, it was 12 months (1.8–23.1) (p = 0.14). Median OS of the patients with the wtKRAS gene was 56.5 months (48.2–64.9), and in the pa- tients with the mutated KRAS gene, it was 58 months (52.6–63.4 months; p = 0.47) (Figure 3). In the patients with the mutation in codon 12, the median OS was 57 months (50.9–64.4), and in the patients with the mutation in codon 13, it was 44 months (40.1–48.4 months; p = 0.40) (Figure 4). Median OS of the patients without mutation in the BRAF gene was 59.2 months (52.5–65.9), and in the patients with the mutated BRAF gene, it was 27.6 months (12.6–42.5), the difference was statistically significant (p = 0.05) (Figure 5). The wtKRAS gene statistically significantly affected the response to the systemic therapy (p = 0.028). Histological pa- rameters included in the analysis and the BRAF gene mutation did not affect significantly the effi- ciency of first line systemic treatment. Of 154 patients, 85 patients were wtKRAS and wtBRAF avaible for additional analysis, for 3 pa- tients there was not possible to determine addition- al biomarker analysis from tumour tissue. In 14 patients (17%) additional mutations in RAS gene were determined. Two patients were NRAS mutated, in 8 patients KRAS codon 61 muta- tion was determined, 4 patients were KRAS codon 146 mutated. Median OS of 68 wtRAS patients was 27 months, median OS of 14 mutated RAS patients was 15 months. Median OS of wtNRAS, codon 61 and FIGURE 3. Overall survival of wtKRAS and mutated KRAS gene patients (p = 0.47). FIGURE 4. Overall survival of mutated KRAS patients in codon 12 and 13 (p = 0.40). Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers90 of the patients treated with chemotherapy and bev- acizumab. Stable disease was achieved in 40% of the patients treated with chemotherapy and cetuxi- mab and in 54% of the patients treated with chemo- therapy and bevacizumab. The disease progressed in 13% of the patients who received chemotherapy and cetuximab, and in 6% of the patients who were treated with chemotherapy and bevacizumab. In the patients with the mutated KRAS gene, the objective response was achieved in 23% of the pa- tients with the mutation in codon 12, and in 47% of the patients with the mutation in codon 13. Stable disease was achieved in 49% of the patients with the mutation in codon 12 and in 41% of the pa- tients with the mutation in codon 13.The disease progressed during first-line treatment in 28% of the patients with the mutation in codon 12 and in 12% of the patients with the mutation in codon 13. Response rate according to systemic treatment and according to mutations in codon 12 and 13 are shown in Table 3 and 4, respectively. Toxicity The most common side effects according to the cri- teria of the National American Institute for Cancer - National Cancer Institute - common toxicity crite- ria (NCI - CTC, version 4.03) were of Grade 1 or 2. The adverse effects of Grade 3 were rare: diarrhea (2%), allergic reaction to cetuximab (2%), rash (1%) and thromboembolic (1% ). Grade 4 adverse reac- tions were diarrhea (1%), leukopenia (1%), neu- FIGURE 5. Overall survival of wtBRAF and mutated BRAF gene patients (p = 0.05). FIGURE 6. Median overall survival of wtNRAS, codon 61 and 146 patients and mutated NRAS, codon 61 and 146 patients (p = 0.072). TABLE 3. Response rate according to systemic treatment Response rate Systemic chemotherapy+cetuksimab, number (%) Systemic chemotherapy+ bevacizumab, number (%) Complete response 7 (13) 6 (17) Partial response 18 (34) 8 (23) Stable disease 21 (40) 19 (54) Progression of disease 7 (13) 2 (6) TABLE 4. Response rate according to mutations in codon 12 and 13 Response rate Codon 12 mutations, number (%) Codon 13 mutations, number (%) Complete response 6 (12) 2 (12) Partial response 5 (11) 6 (35) Stable disease 23 (49) 7 (41) Progression of disease 13 (28) 2 (12) 146 patients was 67.1 months (50.3–67.6), in mu- tated NRAS, codon 61 and 146 patients, it has not been reached yet (p = 0.072) (Figure 6). Median TTP of wtNRAS, codon 61 and 146 patients, was 11.7 months (10.4–14.5), of mtNRAS, codon 61 and 146 patients, was 7.9 months (6.1–11.0 months; p = 0.025) (Figure 7). Efficacy In the patients with the wtKRAS gene, the objective response was achieved in 47% of the patients treat- ed with chemotherapy and cetuximab and in 40% Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers 91 tropenia (1%) and febrile neutropenia (1%). None of the patients died of adverse effects of systemic treatment (Table 5). Discussion To our knowledge a prospective clinical study in patients with CRC determining the prognostic and predictive significance of the histological char- acteristics of primary tumours with mutations in the KRAS and BRAF gene regarding response to treatment has not yet been published. Our study was the first of that kind. The study was conducted from November 2010 to December 2012 and154 patients were included, which was in line with the research plan. The percentage of patients with T4 tumour, G3 tumour, with more than 3 affected re- gional lymph nodes (N2), vascular, perinevral in- vasion and lymphangiosis were comparable with previously published data.2,18 Parameters as perfo- ration of the primary tumour during surgery, pres- ence of tumour deposits in subserosa, mesentery, peritonealised pericolic or perirectal mesenteric tis- sues and iresectability of the primary tumour, were excluded, because of the insufficient number According to our results, poorly differentiated tumour cells and regionally advanced disease with a greater number of affected regional lymph nodes indicate a worse prognosis and a shorter survival of these patients. Other histological parameters of the primary tumour, such as the size of T4 tumour, carcinomatous lymphangiosis, vascular and peri- nevral invasion, which speak of a greater biological aggressiveness of the disease and are important for decision on adjuvant systemic treatment in stages II and III, have not proved to be predictive factors of a worse outcome of the disease, although the difference in size of the primary T4 tumour was marginal (p = 0.08). We hypothesize that the cause of this could be the insufficient number of patients with such histological parameters as well as the short observation period (median observation pe- riod of 15 months), since the median OS in these patients has not yet been reached. Another possible cause may be that these histological parameters are not significantly relevant for the further course of the disease. According to the data from the litera- ture, locally advanced primary tumour, with the infiltration of the free surface of serosa, and the in- cidence of a large number of regional lymph nodes, vascular invasion and carcinomatous lymphangi- osis are independent prognostic factors for worse outcome of the operable disease.2,19 In the case of disseminating disease this parameters also pre- dicts aggressive course although their importance in the spread of disease is not fully defined. The percentage of mutations in the codon 12 and 13 of the KRAS gene was found in 42% of pa- FIGURE 7. Median time to progression of wtNRAS, codon 61 and 146 patients, and mutated NRAS, codon 61 and 146 patients (p = 0.025). TABLE 5. Adverse effects of systemic treatment Grade 1 n (%) Grade 2 n (%) Grade 3 n (%) Grade 4 n (%) Haematological leukopenia neutropenia thrombocytopenia anaemia 42 (27) 42 (27) 33 (21) 88 (57) 8 (5) 8 (5) 11 (7) 10 (6) 0 0 0 0 1 (1) 1+1 (1+1)* 0 0 Non-haematological alopecia fatigue nausea vomiting diarrhea stomatitis hand-foot syndrome peripheral sensory neuropathy hepatic toxicity renal toxicity acneiform rash hypermagnesemia allergic reaction to cetuximab allergic reactions to oxaliplatin arterial hypertension proteinuria bleeding thromboembolic events wound complications perforation arthralgia bladder infection 62 (40) 93 (60) 40 (26) 14 (9) 23 (15) 4 (3) 17 (11) 32 (21) 17 (11) 0 14 (9) 6 (4) 3 (2) 1 (1) 3 (2) 20 (13) 6 (4) 0 0 0 17 (11) / 21 (14) 11 (7) 10 (6) 9 (6) 29 (19) 2 (2) 6 (4) 8 (5) 3 (2) 0 38 (25) 0 1 (1) 0 9 (6) 18 (12) 0 7 (5) 0 0 2 (2) 3 (2) / 0 0 0 2 (2) 0 0 0 0 1 (1) 0 3 (2) 0 0 0 0 1 (1) 0 0 0 0 0 / / / 0 1 (1) 0 / 0 0 0 0 0 0 0 / 0 0 0 / 0 0 0 * 1 patient had febrile neutropenia; / = this grade of adverse effect does not exist; 0 = this grade of adverse effect was not observed in our study Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers92 tients, which is in accordance with the literature data in which 30 to 40% mutations are reported.20-23 According to the data from the retrospective analy- ses of the published clinical studies the mutations in codon 12 are more frequent than mutations in codon 13, which is also confirmed by our study. Namely, 73% of the patients had mutations in co- don 12, and 27% of the patients had mutations in codon 13. All the mutations analyses in our study were performed on the tumour tissue of the pri- mary tumour. Given the proven high degree of compliance of the mutations between tumour tis- sue of the primary tumour and metastases, no fur- ther analysis of the status of mutations in the KRAS gene in operable metastases or their biopsy has been done.24 In our already published retrospec- tive analysis, which included 176 patients, KRAS mutations in the genes were present in 30% of the patients and more often in codon 12 than in codon 13 (84% in codon 12 vs. 16% in codon 13).24 The median OS of patients with the non-mu- tated KRAS gene was shorter in our study than in patients with mutant KRAS genes, but, the dif- ference was not statistically significant (p = 0.47), suggesting that the mutation status in the KRAS gene predicts response to the treatment with EGFR inhibitors, but is not a prognostic factor for sur- vival of the patients. In some patients with a wt- KRAS gene the presence of the V600E mutation in the BRAF gene most likely can be the cause of a shorter survival, since the presence of the V600E mutation in the BRAF gene, according to retrospec- tive reports of previous clinical studies, indicates a worse outcome of metastatic disease.25-28 The mu- tation of the V600E in the BRAF gene was present only in patients with a non-mutated KRAS gene, which was expected in accordance with previous reports of mutually exclusive exclusion of the si- multaneous presence of the both mutations.24 The median OS of patients with mutations in codon 12 was longer than in patients with mutations in co- don 13 but was not significant (p = 0.41). According to the data from the literature, patients with mu- tations in codon 13 have a shorter survival than patients with mutations in codon 12. Patients with mutations in codon 13 respond to treatment with EGFR inhibitors, otherwise in a lower percentage as patients without mutations in the KRAS gene. The retrospective analysis of the efficacy data of these patients with mutations both in codon 12 and codon 13 treated with another EGFR panitumum- ab inhibitor did not confirm this.28-30 In standard treatment with regard to systemic treatment rec- ommendations, patients with mutant KRAS gene do not receive a therapy with EGFR inhibitors.15 In our study, we also analysed the efficacy of sys- temic therapy in patients with mutant KRAS gene with respect to the presence of mutations in codons 12 and 13, none of whom received EGFR therapy. In our analysis, however, we found that patients with mutations in codon 13 respond to a higher percentage to systemic therapy than patients with mutations in codon 12 (47% vs. 23%). The definitive conclusions of this analysis are difficult to give, given that the patients with mutations in codon 13 are in lower number than patients with mutations in codon 12 (27% vs. 73%). These mutations are less frequent in accordance with the already published data from our retrospective study (16% vs. 84%) and other published data (16–28% mutations in the codon 13 KRAS gene).28-30 However, given that they have a shorter survival than patients with muta- tions in codon 12, but respond in a higher percent- age to the first treatment, we believe that results of this analysis will help in the treatment of these patients, and in particular in reducing the disease burden and improving the quality of life. In our study, we also analysed TTP of the dis- ease in patients with operable disease. There were 64 such patients, with operable primary tumour without metastases, 16 patients with rectal cancer and 48 with colon carcinoma. Median time to re- currence of the disease in patients after primary treatment of operable disease with a non-mutated KRAS gene was shorter than in patients with mu- tated KRAS gene (20 vs. 21 months), but the differ- ence was not significant (p = 0.88). We assume that the V600E mutation in the BRAF gene, which was found in four patients with a non-mutated KRAS gene contributes for such results. With this, we confirmed the prognostic importance of the V600E mutation in the BRAF gene for early progression and worse outbreaks in patients with a non-mu- tated KRAS gene.31,32 The TTP of the disease in pa- tients primarily treated for operable disease with operation was longer in those who had mutations in the KRAS codon 12 than in those with mutations in codon 13 (21 vs. 16 months), otherwise the differ- ence was not significant (p = 0.53). Regarding the presence of the V600E mutation in the BRAF gene in our study, as previously stated it was found in four patients, representing 3% of all patients. According to the literature, the frequency of this mutation is from 5 to 10%.2,3,9,13 Also, ac- cording to our previous retrospective analysis, the V600E mutation in the BRAF gene was present at a higher percentage of patients, in 7.4%.24 It is as- sumed that the specificity of the patients involved Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers 93 with such primary tumour characteristics may be due to the fact that the V600E mutation was pre- sent in a lower percentage. All four patients with the V600E mutation had colon cancer, three pa- tients were previously on adjuvant treatment, and one had a primary disseminated disease. In a retrospective analysis of the study involving 94 patients with rectal carcinoma, none of the V600E mutations in the BRAF gene were detected in ei- ther patient.33 This suggests that this mutation is probably less important in the development of rec- tal cancer. The time before progression of the dis- ease in patients after primary treatment is longer for those who did not have the V600E mutation in the BRAF gene than in the V600E mutation (20 to 12 months), but the difference was not significant (p = 0.14). The cause is most likely a small num- ber of patients with this mutation, but such result suggests a worse prognosis of these patients and an earlier recurrence of the disease. In the survival analysis, patients without V600E mutations had a significantly longer median OS than those with V600E mutation (59.2 vs. 27.6 months; p = 0.05). In a retrospective analysis of 176 patients, 13 pa- tients with the V600E mutation in the BRAF gene had a significantly lower expected median survival than patients without this mutation (44.9 vs. 107.4 months; p = 0.04.24 Both studies, both the previous retrospective and the current prospective, confirm the prognostic importance of the V600E mutation in the BRAF gene for predicting a worse disease outcome with shorter patient survival. Furthermore, we examined the response rate ac- cording to the type of systemic treatment in patients with non-mutated KRAS gene. There were 88 such patients. Three hundred and sixty patients with a non-mutated KRAS gene received the first line of systemic treatment with EGFR-inhibitor cetuximab in combination with oxaliplatin or irinotecan and fluoropyrimidines, 35 patients with a non-mutated KRAS gene received the first line of systemic treat- ment with angiogenesis inhibitor bevacizumab in combination with oxaliplatin or irinotecan and fluoropyrimidines. A higher percentage of objec- tive responses (both complete and partial), both for treatment and stability of the disease was achieved in patients with non-mutated KRAS gene receiving cetuximab in combination with oxaliplatin or iri- notecan and fluoropyrimidines. The definitive conclusions about this analysis cannot be made within our research, since this is a small number of patients, and the number of patients was not balanced in number. In addition, the types of systemic chemotherapy that patients received differ. However, it was found that pa- tients with a non-mutated KRAS gene had a better response to systemic first-line treatment involv- ing cetuximab regardless of the type of systemic chemotherapy as patients with the non-mutated KRAS gene that received angiogenesis inhibitor bevacizumab in combination with systemic chem- otherapy. Colorectal adenocarcinoma is a heterogeneous disease. We do not know all the factors that influ- ence the course of the disease and the response to specific systemic therapy. The determination of other mutations in the KRAS gene, such as muta- tions in codon 61 and 146, and mutations in the NRAS gene, has been standard in clinical practice since 2014. The frequency of mutations is about 15% and the patients with a non-mutated RAS gene have about 40% of this mutation. In our retro- spective analysis, we found 17% of these additional mutations, which is consistent with reports from previous literature and they are predictive for re- sponse to EGFR inhibitor therapy. Patients treated with first-line anti-EGFR targeted therapy without these additional mutations in RAS gene, had long- er survival and time to progression compared to the patients who had them. Our results confirmed predictive value of these mutations in RAS gene for anti-EGFR targeted therapy. However, our group of patients is moderate, but it is consisted with pre- viously published data.33,34 Adverse effects of systemic treatment were evaluated according to NCI-CTC, version 4.03. They were mostly of Grade 1 and 2. There were 2% of G3 diarrhea and hypermagnesemia and 1% of G3 bleeding. One patient had G4 adverse effect (febrile neutropenia). In our study, adverse effects were less common than in previous published lit- erature.35-47 None of the patients died of adverse ef- fects. However, all patients with RAS non-mutated genes do not respond to treatment with anti-EG- FR inhibitors. Other biomarkers have to be de- termined. One of them is mismatch repair- defi- ciency (MSI-H), which is present in 3.5% to 5.0% of mCRC.48,49 Immunotherapy with checkpoint inhibitors is recommended in second-line therapy in these patients. In clinical trials, there are other biomarkers in response to specific systemic thera- py, such as amfiregulin and epiregulin, PI3K and PTEN mutations, but are still not recommended in regular clinical practice.50 In order to made definite reasoning on this issue we need more information and more studies, both, retrospective and prospec- tive should be performed. Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers94 Conclusions According to the results of our prospective study G3 of tumour differentiation (poorly differentiated adenocarcinoma) and N2 status of regional lymph nodes (more than 3 metastatic regional lymph nodes) show a greater biological aggressiveness of the disease and shorter survival after the first re- lapse, but do not affect the efficacy of the first-line systemic therapy. The mutations in the KRAS gene have predic- tive value for the response to treatment with EGFR inhibitors, but do not have prognostic significance for the survival of patients with disseminated colo- rectal adenocarcinoma. The KRAS mutations in codon 12 and 13 differently affect the survival of these patients and the response rate of first-line systemic therapy. The V600E mutation in the BRAF gene does not have the prognostic significance for the decision on the type of first-line systemic therapy, but it pre- dicts a poor prognosis of the disease and shorter survival of these patients at first relapse. According to our restrospective analysis of ad- ditional mutations in NRAS gene and in KRAS co- don 61 and 146 are predictive for treatment with anti-EGFR targeted therapy. Acknowledgement The research was founded by Slovenian Research Ageny (ARRS), grant number P3-0321. References 1. Cancer in Slovenia 2015. Ljubljana: Institute of Oncology Ljubljana, Epidemiology and Cancer Registry, Cancer Registry of republic of Slovenia; 2018. 2. Schmoll1 HJ, Van Cutsem E, Stein A, Valentini V, Glimelius B, Haustermans K, et al. ESMO consensus guidelines for management of patients with colon and rectal cancer. A personalized approach to clinical decision making. Ann Oncol 2012; 23: 2479-516. doi: 10.1093/annonc/mds236 3. Van Cutsem1 E, Cervantes A, Adam R, Sobrero A, Van Krieken JH, Aderka D, et al ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol 2016; 1386-422. doi: 10.1093/ annonc/mdw235 4. National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: colon cancer. V.2.2018. [cited 2018 Jun 15]. Available at: http:// www.nccn.org/profesionals/phisician_gls/PDF/colon.pdf 5. National Comprehensive Cancer Network: NCCN clinical practice guidelines in oncology: rectal cancer. V.2.2018 [cited 2018 Aug 3]. Available at: http:// www.nccn.org/profesionals/phisician_gls/PDF/colon.pdf 6. Biagi JJ, Raphael MJ, Mackillop WJ, Kong W, King WD, Booth CM. Association between time to initiation of adjuvant chemotherapy and survival in colo- rectal cancer. A systemic review and meta-analysis. JAMA 2011; 305: 2335- 42. doi: 10.1001/jama.2011.749 7. Ocvirk J. Advances in the treatment of metastatic colorectal carcinoma. Radiol Oncol 2009; 43: 1-8. doi: 10.2478/v10019-009-0004-1 8. Meyerhardt JA, Mayer RJ. Systemic therapy for colorectal cancer. N Engl Med 2005; 352: 476-87. doi: 10.1053/j.gastro.2008.02.098 9. Esin E, Yalcin S. Maintenance strategy in metastatic colorectal cancer: a systematic review. Cancer Treat Rev 2016; 42: 82-90. doi: 10.1016/j. ctrv.2015.10.012 10. Advanced colorectal cancer: ESMO clinical practice guidelines for diagnosis, treatment. Ann Oncol 2010; 21: v93- 7. doi: 10.1093/annonc/mdq222 11. Markowitz SD, Bertagnolli MM. Molecular basis of colorectal cancer. N Engl J Med 2009; 361: 2449-60. doi: 10.1056/NEJMra0804588 12. De Roock W, Jonker DJ, Di Nicolantonio F, Sartore-Bianchi A, Tu D, Siena S, Lamba S, et al. Association of KRAS p.G13D mutation with outcome in patients with chemotherapy-refractory metastatic colorectal cancer treated with cetuximab. JAMA 2010; 304: 1812- 20. doi: 10.1001/jama.2010.1535 13. De Roock W, Claes B, Bernasconi D, De Schutter J, Biesmans B, Fountzilas G, et al. Effect of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. Lancet Oncol 2010; 11: 753-62. doi: 10.1016/S1470-2045(10)70130-3 14. Tol J, Nagtegaal ID, Punt CJA. BRAF mutation in metastatic colorectal cancer. N Engl J Med 2009; 361: 98-99. doi: 10.3978/j.issn.2078-6891.2015.077 15. Samowitz WS, Sweeney C, Herrick J, Albertsen H, Levin TR, Murtaugh MA, et al. Poor survival associated with the BRAF V600E mutation in microsatel- lite- stable colon cancer. Caner Res 2005; 65: 6063-9. doi: 10.1158/0008- 5472.CAN-05-0404 16. Ličar A, Cerkovnik P, Novaković S. Distribution of some activating KRAS and BRAF mutations in Slovene patients with colorectal cancer. Med Oncol 2011; 28: 1048-53. doi: 10.1007/s12032-010-9631-z 17. Fariña-Sarasqueta A, van Lijnschoten G, Moerland E, Creemers GJ, Lemmens VE, Rutten HJ, et al. The BRAF V600E mutation is an independent prognostic factor for survival in stage II and stage III colon cancer patients. Ann Oncol 2010; 21: 2396-402. doi: 10.1093/annonc/mdq258 18. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumors: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: 228-47. doi: 10.1016/j.ejca.2008.10.026 19. Marzouk O, Schofield J. Review of histopathological and molecular prog- nostic features in colorectal cancer. Cancers 2011; 3: 276-81. doi: 10.3390/ cancers3022767 20. Ličar A, Cerkovnik P, Ocvirk J, Novaković S. KRAS mutations in Slovene pa- tients with colorectal cancer: frequency, distribution and correlation with the response to treatment. Int J Oncol 2010; 36: 1137-44. doi: 10.3892/ ijo_00000596. 21. Artale S, Sartore- Bianchi A, Veronese SM, Gambi V, Sarnataro CS, Gambacorta M, et al. Mutations of KRAS and BRAF in primary and matched metastatic sites of colorectal cancer. J Clin Oncol 2008; 26: 4217- 9. doi: 10.1200/JCO.2008.18.7266 22. Bokemeyer C, Van Cutsem E, Rougier P, Ciardiello F, Heeger S, Schlichting M, et al. Addition of cetuximab to chemotherapy as first-line treatment for KRAS wild-type metastatic colorectal cancer: pooled analysis of the CRYSTAL and OPUS randomised clinical trials. Eur J Cancer 2012; 48: 1466- 75. doi: 10.1016/j.ejca.2012.02.057 23. Artale S, Sartore- Bianchi A, Veronese SM, Gambi V, Sarnataro CS, Gambacorta M, et al. Mutations of KRAS and BRAF in primary and matched metastatic sites of colorectal cancer. J Clin Oncol 2008; 26: 4217-9. doi: 10.1200/JCO.2008.18.7286 24. Rebersek M, Boc M, Škerl P, Benedik J, Hlebanja Z, Volk N, et al. Efficacy of first- line systemic treatment in correlation with BRAF V600E and different KRAS mutations colorectal cancer - a single institution retrospective analy- sis. Radiol Oncol 2011; 45: 285-91. doi: 10.2478/v10019-011-0039-y 25. Lambrechts D, De Roock W, Prenen H, De Schutter J, Jacobs B, Biesmans B, et al. The role of KRAS, BRAF, NRAS and PIK3CA mutations as a markers of resistance of cetuximab in chemorefractory metastatic colorectal cancer. J Clin Oncol 2009; 27: 9020. doi: 10.1158/1078-0432.CCR-08-2961 26. Tejpar S, De Roock W. PIK3CA, BRAF and KRAS mutations and outcome prediction in chemorefractory metastatic colorectal cancer (mCRC) patients treated With EGFR targeting monoclonal antibodies (MoAbs): results of a European Consortium.[abstract]. Eur J Cancer Suppl 2009; 7: 322. No. 6005 Radiol Oncol 2019; 53(1): 85-95. Rebersek M et al. / Metastatic colorectal cancer and biomarkers 95 27. Loupakis F, Ruzzo A, Cremolini C, Vincenzi B, Salvatore L, Santini D, et al. KRAS codon 61, 146 and BRAF mutations predict resistence to cetuximab plus irinotecan in KRAS codon 12 and 13 wild-type metastatic colorectal cancer. Br J Cancer 2009; 101: 715-21. doi: 10.18632/oncotarget.13697 28. Tejpar S, Celik I, Schlichting M, Sartorius U, Bokemeyer C, Van Cutsem E. Association of KRAS G13D tumor mutations with outcome in patients with metastatic colorectal cancer treated with first-line chemotherapy with or without cetuximab. J Clin Oncol 2012; 30: 3570-7. doi: 10.1200/ JCO.2012.42.2592 29. Peeters M, Douillard JY, Van Cutsem E, Siena S, Zhang K, Williams R, et al. Mutant KRAS codon 12 and 13 alleles in patients with metastatic colorectal cancer: assessment as prognostic and predictive biomarkers of response to panitumumab. J Clin Oncol 2013; 31: 759-65. doi: 10.1200/ JCO.2012.45.1492. Epub 2012 Nov 26. 30. Roth AD, Tejpar S, Dolorenzi M, Yan P, Fiocca R, Klingbiel D, et al. Prognostic role of KRAS and BRAF in stage II and III resected colon cancer: results of the translational study on the PETACC-3, EORTC 40993, SAKK 60-00 trial. J Clin Oncol 2010: 28: 466-74. doi: 10.1200/JCO.2009.23.3452. 31. Fujiyoshi K, Yamamoto G, Takahashi A, Arai Y, Yamada M, Kakuta M, et al. High concordance rate of KRAS/BRAF mutations and MSI-H between pri- mary colorectal cancer and corresponding metastases. Oncol Rep 2017; 37: 785-92. doi: 10.3892/or.2016.5323. 32. Gaedcke J, Grade M, Jung K, Schirmer M, Jo P, Obermeyer C, et al. KRAS and BRAF mutations in patients with rectal cancer treated with preopera- tive chemoradiotherapy. Radiother Oncol 2010; 94: 76-81. doi: 10.1016/j. radonc.2009.10.001 33. Pietrantonio F, Cremolini C, Petrelli F, Di Bartolomeo M, Loupakis F, Maggi C, et al. First-line anti-EGFR monoclonal antibodies in panRAS wild-type metastatic colorectal cancer: a systematic review and meta-analysis. Crit Rev Oncol Hematol 2015; 96: 156-66. doi: 10.1016/j.critrevonc.2015.05.016 34. Sorich MJ, Wiese MD, Rowland A, Kichenadasse G, McKinnon RA, Karapetis CS, et al. Extended RAS mutations and anti-EGFR monoclonal antibody sur- vival benefit in metastatic colorectal cancer: a meta-analysis of randomized, controlled trials. Ann Oncol 2015; 26: 13-21. doi: 10.1093/annonc/mdu378 35. Scartozzi M, Galizia E, Chiorrini S, Giampieri G, Berardi R, Pieratoni C, et al. Arterial hypertension correlates with clinical outcome in colorectal cancer treated with first- line bevacizumab. Clin Oncol 2009; 20: 227-30. doi: 10.1093/annonc/mdn637 36. Mir O, Ropert S, Alexandre J, Goldwasser F. Hypertension as a surrogate marker for the activity of anti- VEGF agents. Ann Oncol 2009; 20: 967-70. doi: 10.1093/annonc/mdp206 37. Poole C, Gardiner J, Twelves C, Johnston P, Harper P, Cassidy J, et al. Effect of renal impairment on the pharmacokinetics and tolarebility of capecitabine (Xeloda) in cancer patients. Cancer Chemother Pharmacol 2002; 49: 225-34. doi: 10.1007/s00280-001-4808-0 38. Cassidy J, Twelves C, Van Cutsem E, Hoff P, Bajetta E, Boyer M, et al. First-line oral capecitabine therapy in metastatic colorectal cancer: a favorable safety profile compared with intravenous 5-fluorouracil/leucovorin. Ann Oncol 2002; 13: 566-75. doi: 10.1200/JCO.2001.19.21.4097 39. Hoskins JM, Goldberg RM, Qu P, Ibrahim JG, McLeod HL. UGT1A1* 28 geno- type and irinotecan- induced neutropenia: dose matters. J Nati Cancer Inst 2007; 99: 1290-421. doi: 10.1200/JCO.2009.21.7943 40. Hoskins JM, McLeod HL. UGT1A and irinotecan toxicity: keeping it in the family. J Clin Oncol 2009; 27: 2419-21. doi: 10.2217/pgs.10.95 41. Pereg D, Lishner M. Bevacizumab treatment for cancer treatment with car- diovascular disease: a double edged sword? Eur Heart J 2008; 29: 2325-26. doi: 10.1093/eurheartj/ehn384. 42. Scappaticci FA, Skillings JR, Holden SN, Gerber HP, Miller K, Kabbinavar F, et al. Arterial thromboembolic events in patients with metastatic colorectal cancer treated with chemotherapy and bevacizumab. J Nati Cancer Inst 2007; 1232-9. doi: 0.1093/jnci/djm086 43. Lee CS, Ryan EJ, Glen A Doherty GA. Gastro-intestinal toxicity of chemo- therapeutics in colorectal cancer: the role of inflammation. World J Gastroenterol 2014; 20: 3751-61. doi: 10.3748/wjg.v20.i14.3751 44. Ocvirk J, Rebersek M. Management of cutaneous side effect of cetuximab therapy with K1 creme. Radiol Oncol 2008; 42: 215-24. doi: 10.2478/ v10019-008-0022-4 45. Ocvirk J, Cencelj S. Management of cutaneous side-effect of cetuximab therapy in patients with metastatic colorectal cancer. JEADV 2010; 24: 453- 9. doi: 10.1111/j.1468-3083.2009.03446.x 46. Ocvirk J, Rebersek M, Boc M, Mesti T, Ebert Moltara M. Prophylactic use of K1 crem for reducing skin toxicity during cetuximab treatment in patients with metastatic colorectal cancer (mCRC). J Clin Oncol 2016; 28(15 Suppl): V124. doi: 10.1200/jco.2010.28.15_suppl.e14011. 47. Siena S, Glynne-Jones R, Adenis A, Thaler J, Preusser P, Aguilar EA, et al. Reduced incidence of infusion- related reactions in metastatic colorectal cancer during treatment with cetuximab plus irinotecan with combined corticosteroid and antihistamin premedication. Cancer 2010, 116: 1827-37. doi: 10.1002/cncr.24945 48. Venderbosch, ID Nagtegaal, TS Maughan, CG Smith , JP Cheadle, D Fisher, et al. Mismatch repair status and BRAF mutation status in metastatic colo- rectal cancer patients: a pooled analysis of the CAIRO, CAIRO2, COIN, and FOCUS studies. Clin Cancer Res 2014 20: 5322-30. doi: 10.1158/1078-0432. CCR-14-0332 49. Le DT, Uram JN, Wang H, Bartlett BR, Kemberling H, Eyring AD, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med 2015; 372: 2509-20. doi: 10.1056/NEJMoa1500596 50. Therkildsen C, Bergmann TK, Henrichsen-Schnack T, Ladelund S, Nilbert M. The predictive value of KRAS, NRAS, BRAF, PIK3CA and PTEN for anti- EGFR treatment in metastatic colorectal cancer: a systematic review and meta-analysis. Acta Oncol 2014; 53: 852-64. doi: 10.3109/0284186X.201 4.895036