IHC was performed routinely and was successful in all 109 cases. Routine mutational analyses were requested for KRAS (76), NRAS (27), and BRAF (56) and all carried out successfully. Retrospective NGS analysis was performed on 93 cases. The data obtained from resection specimens were comparable to that obtained from the respective endoscopic biopsy with the exception of PTEN, C-MET, and HER-2 (see below).
MMR Status and BRAF V600E Mutation Status
There was complete concordance between endoscopic biopsy and resection specimen for MMR marker IHC (MSH2, MSH6, MLH1, and PMS2). All 4 MMR proteins were expressed in 93 cases, and classified as proficient for MMR (pMMR). In 16 cases, ≥1 MMR proteins were abnormal, that is, absent, and were classified as deficient for MMR (dMMR). Four of these had wild-type BRAF, raising the possibility of Lynch syndrome (LS). One had loss of MSH6 whereas the other 3 had loss of both MLH1 and PMS2.
BRAF mutation status was evaluated by IHC for all 109 cases. We verified these results in 101 cases using PCR only (n=8), NGS only (n=45), or both (n=48). We were unable to verify the IHC result for the remaining 8 cases because of repeated failure of NGS testing. IHC-based BRAF mutation status was concordant with NGS & PCR findings except for mutation c.1816G>A (p.Gly606Arg), which was identified with NGS. This is a Tier 3 mutation, of indeterminate significance, distal to codon 600, and was unsurprisingly not picked up by either PCR or IHC. Such cases were therefore deemed nonmutated. There was complete concordance between endoscopic biopsies and resection specimens for BRAF IHC.
KRAS and NRAS Mutation Status
KRAS mutation status was determined successfully for a total of 105 cases. It was determined by PCR or pyrosequencing only (n=12), NGS only (n=29), or both (n=64). There were 2 cases where a mutation identified by NGS was not picked up by PCR (KRAS c.38G>A [p.Gly13Asp] and KRAS c.183A>C [p.Gln61His]). Both are Tier 1 mutations and therefore included as KRAS mutations within this study.
NRAS mutation status was determined successfully in a total of 98 cases by PCR or pyrosequencing only (n=5), by NGS only (n=71) or by both methods (n=22) with full concordance.
PTEN Loss, HER-2, and c-MET Over Expression
Where available, these biomarkers were scored on the endoscopic biopsy as well as the resection specimen. There was discordance in PTEN expression between biopsy and resection in 8.3% of cases (7/84). In all discordant cases PTEN was present in the endoscopic biopsy but lost in the resection. Discordant c-MET expression occurred in 25% of cases with biopsy (7/28). These were not the same 7 cases as above. In 6 cases there was a decreased expression in the resection. Discordant HER-2 expression occurred in just 1 case (of 80 with biopsy), where focal 3+ expression was not seen in the biopsy.
We wanted to understand if a DGH IHC laboratory, with support from a reference center, can carry out up-front/reflex testing to stratify CRC within the timeframe required for informing clinical decision making and we wanted to provide a framework for the production and presentation of this molecular pathology data. We evaluated whether it is possible to perform the selected tests (MMR status, BRAF mutation status, RAS mutation status, loss of PTEN, and overexpression of c-MET and HER-2) within a 2-week turnaround time (TAT) and in time for the local MDT.
We demonstrated that all slide-based tests were easily performed in-house within a 48-hour TAT. The mutation analyses were sent off-site to a reference center as part of the routine diagnostic pathway and had a TAT of 8 to 10 working days. It was therefore only possible to collate all the results in time for MDT if a suitable block was selected at the time of cut-up and sent away promptly, without waiting for the diagnostic report. If TAT of referral centers cannot be improved, the prompt dispatch of the tissue block is critical and needs to be inserted into the routine cut-up procedure. A recent study has shown that using a NGS panel approach achieves a median turn-around time of 7 days at a cost which is increasingly competitive compared to single gene testing as more targets are added.10
MMR Deficiency and LS
Recent NICE guidance states that all CRC patients, regardless of their age, should have tumor-based testing to assess the risk of LS when first diagnosed.6 LS is the most common cause of hereditary bowel cancer and carries an increased risk of developing other cancers.11 LS is estimated to cause 1000 cases of bowel cancer each year in UK, yet fewer than 5% of people with this condition are currently identified.12
This guidance significantly increases the amount of testing required but our study demonstrates that testing all CRC patients for MMR using in-house IHC in a DGH is feasible. The addition of BRAF IHC allows the identification of BRAF V600E mutation-negative patients who require referral to genetic services for further investigations for LS. These IHC tests can be performed on either the endoscopic biopsy or resection specimen, as we demonstrated 100% concordance. NICE guidelines recommend that patients negative for both BRAF V600E mutation and MLH1 require an MLH1 promoter hypermethylation test.6 Only 3 patients in our cohort (2.8%) fell into this category, demonstrating that in-house IHC would be sufficient for the majority of cases.
MMR testing by IHC can be done quickly and reliably before treatment in order to support clinical decision making as patients with dMMR tumors may have better prognosis,13 may not benefit from adjuvant chemotherapy,14 may benefit from low dose aspirin15 and respond to immunomodulation through checkpoint inhibitors.16 The FDA has recently granted accelerated approval to Pembrolizumab in certain situations for patients with any type of dMMR solid tumor,17 emphasizing the importance of universal MMR testing.
Implications for Targeted/Biological Therapies
Prompt identification of dMMR or LS patients is only one aspect of CRC biomarking. It is a rapidly growing and constantly evolving area, but we have demonstrated that a DGH can implement the necessary service improvements to take advantage of new biomarkers and provide high quality testing with adequate turnaround time for patient treatment in a local setting.
The targeting of the EGF-R signaling pathway is a major therapeutic option in CRC (Fig. 1) and the regulatory approval for drugs targeting this pathway is dependent on absence of activating mutations in the RAS genes. Although RAS mutation status was only immediately relevant for 4 of our patients (those presenting with advanced/metastatic disease), an estimated recurrence rate of 20% to 30% for stage II and 50% to 80% for stage III patients18 means that 75% of our cohort would need this data to inform treatment in the near future. This would provide ample justification for immediate reflex testing rather than on-demand at a later date. In fact, immediate reflex testing provides higher quality information in the pathology report that is, most importantly, rapidly accessible upon recurrence. In addition, as we have shown in the prostate setting, reflex testing allows more effective use of service resources thus paradoxically creating capacity (manuscript in preparation).
Evidence suggests that changes in many other molecules along the EGF-R signaling pathways may impact on response to inhibitors of these pathways. For example, the presence of BRAF activating mutations affect the response to EGF-R inhibitors such as cetuximab and panitumumab.19 Likewise, loss of PTEN, over-expression of c-MET or HER-2 have a negative effect on response to EGF-R inhibitors although there is some controversial literature.8,20–22 Certainly, the current selection criteria for EGF-R TKI therapy results in a number of treatment failures, suggesting that refinements in the selection are necessary.23
Since ours was a feasibility study, the MDT did not act upon our additional test results. Nevertheless, we retrospectively evaluated their effect on eligibility for EGF-R targeted therapy. According to current guidelines, 66% of our patients would be eligible for these drugs. The addition of BRAF mutation status would bring this down to 45%. The inclusion of PTEN and HER-2 would reduce this cohort to 31% and, using all our data, we would predict that only 15 patients (14%) would respond optimally. These tests therefore may have huge implications on treatment decisions. Although we used published scoring systems for PTEN, c-MET, and HER-2 (Table 1), there are consensus issues,22–27 so it is crucial that suitable scoring systems are devised and validated against clinical response for these markers, and an external quality assurance process is established. We found disparities between scores for resection and biopsy tissue, which may be because of fixation/preanalytical processing or a reflection of disease process.
Other potential targeted therapies may be beneficial to CRC patients. For example, HER-2 overexpression may indicate good response to trastuzumab and lapatinib (Heracles trial28), loss of PTEN may indicate good response to mTOR inhibitors, which target the AKT pathway downstream of PTEN29 and over expression of c-MET may indicate response to MET and MEK inhibitors (MErCuRIC1 trial30). The FOCUS4 trial is currently stratifying CRC using biomarkers such as BRAF, PIK3/PTEN, and RAS to inform treatment.31 Our study complements this trial by demonstrating the feasibility of using these tests for routine stratification in a small DGH.
Development of the Algorithm
Many of the authors (C.D., T.T., I.M.F., K.M., and A.C.) have had long associations with MDT/TB and understand the challenges of presenting ever-increasing molecular pathology data with complex ramifications for treatment decisions. The MDT/TB has only a few minutes allocated to each patient so requires a system of communication that summarizes all findings, is easily and quickly interpretable and can guide clinical decisions. We therefore constructed a graphic representation, structured to follow the then current clinical decision-making. Since then, rapid progress in CRC biomarker research has impacted further on treatment decisions. For instance, immunoscore,32 TILs (tumor infiltrating lymphocytes) and PD-L1 assessment were not widely used at the time of the study and immunomodulation with checkpoint inhibitors was unavailable as a treatment option in 2013.
We have therefore expanded and revised the algorithm to reflect what would be a working classification in 2017 and included test results with prognostic and therapeutic implications as well as traditional anatomical data (Fig. 4). This new algorithm is easy to update as new biomarkers emerge and guidelines change and can be customized to reflect local oncological practice.
NICE concluded that testing using IHC for MMR plus BRAF and MLH1 promoter methylation is a cost-effective use of NHS resources.6,33,34 Our approach to CRC testing could easily be adopted by all hospitals in response to NICE guidelines. In its most succinct form, this can be done with 5 IHC tests (MLH1, PMS2, MSH2, MSH6, and BRAF), with a relatively low burden on resources, although the health-economic case is predicated on central funding given the benefits to the NHS lie outside of pathology budgets.33 Additional tests could provide more accurate prediction and prognosis, thus reducing costs and delay caused by ineffective treatment.35
In the UK there is a large gap between the provision of cancer testing and demand.36 The estimated gap in CRC is the largest and in 2014 affected 10,704 patients (49%) who did not receive testing, potentially missing out on optimal treatment. The UK commissioning system funds tests for systemic anticancer treatment centrally whereas traditional IHC tests are commissioned locally. Funding of some IHC tests has therefore become separated from the funding of the associated targeted treatment. Furthermore, NHS departmental budgets are compartmentalized, so that savings made in Oncology and Surgery through the improvement of outcomes by personalized medicine are unavailable to Pathology. This is serious, as the health-economic case is predicated on central funding, given the benefits to the NHS lie outside of Pathology.33 Such current structures do not therefore fit current requirements and threaten the implementation of improved care pathways with a proven health-economic basis.
The use of IHC for drug selection for breast and lung cancer is well established.37,38 When other modalities of testing have been available, IHC is the most efficient and cost-effective platform.38 Our study demonstrates that IHC has a significant role to play in personalized medicine for CRC. The recent introduction of PD-L1 testing (the biomarker for checkpoint inhibitor therapy that can only be done by IHC in Histopathology services) reinforces its importance.
It is clear that we need a unified strategy to fund all companion diagnostics, irrespective of whether they are test tube-based or slide-based. If we fail to finance appropriately and adequately all tests that allow patients to receive the most appropriate medicine, we fail both patients and all who pay into the health system, as well as making a mockery of health-economic studies.
We hope this work will enable pathologists to take up the challenges of supporting personalized cancer treatment whether they work in large centers or in small hospitals. We acknowledge that the future is moving toward screening large panels of biomarkers and this may even involve liquid biopsies as opposed to tissue biopsies. Once such systems are established, running in sufficient quantities and can demonstrate concordance and quality, they may well be the most effective way of determining choice of targeted therapy. We are not there yet. We have to fill the gap for the patients of today. This requires education, an understanding of the current limitations as well as the future possibilities and the development of funding streams which are not divisive.
The authors acknowledge the support of Roche Tissue Diagnostics for funding this work and the Technology Strategy Board for funding the NGS panel. We are indebted to the following for advice, discussion and help with manuscript preparation: Anne Waydelich, Gilles Erb, Uwe Schalles, Paula Toro, Joakim Jagorstrand and Chris Hudson.
1. Global Burden of Disease Cancer Collaboration. The global burden of cancer 2013. JAMA Oncol. 2015;1:505–527.
2. Birkenkamp-Demtroder K, Olesen SH, Sørensen FB, et al. Differential gene expression in colon cancer of the caecum versus the sigmoid and rectosigmoid. Gut. 2005;54:374–384.
3. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer
. N Engl J Med. 2008;359:1757–1765.
4. Amado RG, Wolf M, Peeters M, et al. Wild type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer
. J Clin Oncol. 2008;26:1626–1634.
6. Molecular testing strategies for Lynch syndrome
in people with colorectal cancer
. Diagnostics guidance [DG27]. National Institute for Health and Care Excellence. Diagnostics guidance Published: 22 February 2017. Available at: https://www.nice.org.uk/guidance/dg27
Accessed April 12, 2017.
8. Lo Nigro C, Ricci V, Vivenza D. Prognostic and predictive biomarkers in metastatic colorectal cancer
anti-EGFR therapy. World J Gastroenterol. 2016;22:6944–6954.
9. Bardelli A, Corso S, Bertotti A, et al. Amplification of the MET receptor drives resistance to anti-EGFR therapies in colorectal cancer
. Cancer Discov. 2013;3:658–673.
10. Hamblin A, Wordsworth S, Fermont JM, et al. Clinical applicability and cost of a 46-gene panel for genomic analysis of solid tumours: Retrospective validation and prospective audit in the UK National Health Service. PLoS Med. 2017;14:e1002230.
11. Møller P, Seppälä T, Bernstein I, et al. Cancer incidence and survival in Lynch syndrome
patients receiving colonoscopic and gynaecological surveillance: first report from the prospective Lynch syndrome
database. Gut. 2017;66:464–472.
12. Adelson M, Pannick S, East JE, et al. UK Colorectal cancer
patients are inadequately assessed for Lynch syndrome
. Frontline Gastroenterol. 2014;5:31–35.
13. Popat S, Hubner R, Houlston RS, et al. Systematic review of microsatellite instability and colorectal cancer
prognosis. J Clin Oncol. 2005;23:609–618.
14. Sargent DJ, Marsoni S, Monges G, et al. Defective mismatch repair as a predictive marker for lackof efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol. 2010;28:3219–3226.
15. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch repair deficiency. N Engl J Med. 2015;372:2509–2520.
16. Burn J, Gerdes A-M, Macrae F, et al. Long-term effect of aspirin on cancer risk in carriers of hereditary colorectal cancer
: an analysis from the CAPP2 randomised controlled trial. Lancet. 2011;378:2081–2087.
18. Kanwar SS, Poolla A, Majumdar APN, et al. Regulation of colon cancer recurrence and development of therapeutic strategies. World J Gastrointest Pathophysiol. 2012;3:1–9.
19. Di Nicolantonio F, Martini M, Molinari F, et al. Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer
. J Clin Oncol. 2008;26:5705–5712.
20. Therkildsen C, Bergmann TK, Henrichsen-Schnack T, et al. 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;53:852–864.
21. De Stefano A, Carlomagno C. Beyond KRAS: Predictive factors of the efficacy of anti-EGFR monoclonal antibodies in the treatment of metastatic colorectal cancer
. World J Gastroenterol. 2014;20:9732–9743.
22. Molinari F, Frattini M. Functions and regulation of the PTEN gene in colorectal cancer
. Front Oncol. 2014;3:326.
23. Miyamoto Y, Suyama K, Baba H. Recent advances in targeting the EGFR signaling pathway for the treatment of metastatic colorectal cancer
. Int J Mol Sci. 2017;18:752.
24. Liu Y, Yu X-F, Zou J, et al. Prognostic value of c-Met in colorectal cancer
: a meta-analysis. World J Gastroenterol. 2015;21:3706–3710.
25. Bradley CA, Dunne PD, Bingham V, et al. Transcriptional upregulation of c-MET is associated with invasion and tumor budding in colorectal cancer
. Oncotarget. 2016;7:78932–78945.
26. Valtorta E, Martino C, Sartore-Bianchi A, et al. Assessment of a HER2 scoring system for colorectal cancer
: results from a validation study. Mod Pathol. 2015;28:1481–1491.
27. Blok EJ, Kuppen PJK, van Leeuwen JEM, et al. Cytoplasmic overexpression of HER2: a key factor in colorectal cancer
. Clin Med Insights: Oncol. 2013;7:41–51.
28. Sartore-Bianchi A, Trusolino L, Martino C, et al. Dual-targeted therapy
with trastuzumab and lapatinib in treatment-refractory, KRAS codon 12/13 wild-type, HER2-positive metastatic colorectal cancer
(HERACLES): a proof-of-concept, multicentre, open-label, phase 2 trial. Lancet Oncology. 2016;17:738–746.
29. Janku F, Hong DS, Fu S, et al. PIK3CA, and PTEN aberrations in early-phase trials with PI3K/AKT/mTOR Inhibitors: experience with 1,656 patients at MD Anderson Cancer Center. Cell Rep. 2014;6:377–387.
30. Van Schaeybroeck S, Rolfo CD, Élez E, et al. MErCuRIC1: A Phase I study of MEK1/2 inhibitor PD-0325901 with cMET inhibitor crizotinib in RASMT and RASWT (with aberrant c-MET) metastatic colorectal cancer
(mCRC) patients. JCO 2015 ASCO Annual meeting, Vol 33, no 15_suppl. Available at: http://meetinglibrary.asco.org/content/151724-156
. Accessed May 3, 2017.
31. Richman SD, Adams R, Quirke P, et al. Pre-trial inter-laboratory analytical validation of the FOCUS4 personalised therapy trial. J Clin Pathol. 2016;69:35–41.
32. Mlecnik B, Bindea G, Angell HK, et al. Integrative analyses of colorectal cancer
show immunoscore is a stronger predictor of patient survival than microsatellite instability. Immunity
33. Snowsill T, Huxley N, Hoyle M, et al. A systematic review and economic evaluation of diagnostic strategies for Lynch syndrome
. Health Technol Assess. 2014;18:1–406.
34. Snowsill T, Huxley N, Hoyle M, et al. A model-based assessment of the cost–utility of strategies to identify Lynch syndrome
in early-onset colorectal cancer
patients. BMC Cancer. 2015;15:313.
37. Dolan M, Snover D. Comparison of immunohistochemical and fluorescence in situ hybridization assessment of HER-2 status in routine practice. Am J Clin Pathol. 2005;123:766–770.
38. Doshi S, Ray D, Stein K, et al. Economic analysis of alternative strategies for detection of ALK rearrangements in non small cell lung cancer. Diagnostics. 2016;6:4.
39. Sangale Z, Prass C, Carlson A, et al. A robust immunohistochemical assay for detecting PTEN expression in human tumors. Appl Immunohistochem Mol Morphol. 2011;19:173–183.
40. Seo AN, Kwak Y, Kim DW, et al. HER2 status in colorectal cancer
: its clinical significance and the relationship between HER2gene amplification and expression. PLoS One. 2014;9:e98528.
Keywords:Copyright 2019 Wolters Kluwer Health, Inc. All rights reserved.
colorectal cancer; Lynch syndrome; mismatch repair proteins; personalized medicine; targeted therapy