Abstract
The implementation of stratified medicine in modern cancer care presents substantial opportunity to refine diagnosis and treatment but also numerous challenges. Through experience in a UK tumor profiling initiative, we have gained valuable insights into the complexities and possible solutions for routine delivery of stratified cancer medicine.
Genetic discovery in cancer medicine has set the narrative for clinical oncology progress in the 21st century. BRAF was identified as an oncogene in 2002, with its expressed protein subsequently targeted with great clinical success in BRAF-mutated melanoma Citation[1,2]. Concepts, such as oncogene addiction and synthetic lethality, have helped us to develop and apply targeted agents, including imatinib in chronic myeloid leukemia and poly-ADP-ribose polymerase inhibition in BRCA-mutated breast and ovarian cancer Citation[3,4]. Even previous areas of deficit, such as the role of epidermal growth factor receptor (EGFR) in non-small cell lung cancer, have become more defined, with early results from the latest generation of inhibitors, suggesting erlotinib and gefitinib resistance may soon be overcome in EGFR-mutated disease Citation[5–7].
In addition to the potentially transformative opportunities and numerous successes of applying stratified medicine to cancer care, the experience of the past decade has highlighted numerous challenges. Genetic technology has developed at a rapid pace, but in parallel so has the trend for obtaining smaller diagnostic tissue samples through less invasive methods, meaning that as the scope of analysis increases, the quantity and quality of formalin-degraded archival nucleic acids are in danger of becoming the rate-limiting step. In addition, several recent and ongoing studies have explored intratumoral heterogeneity and clonal evolution, representing a challenge for molecular characterization and an impediment to sustained treatment responses Citation[8].
The Cancer Research UK Stratified Medicine Programme (SMP) was conceived and designed in response to an increasing demand for early analysis of multiple prognostic and predictive genetic markers in clinical tumor samples Citation[9]. With whole-genome level initiatives such as The Cancer Genome Atlas and International Cancer Genome Consortium delineating the full range of cancer genomics, it is highly likely that further progress will follow, building on the success of those described above. The need for initiatives, such as this, to establish the expertise and infrastructure for molecular diagnostics in solid tumors is exemplified by the following two key clinical deficits observed in the UK’s National Health Service at the outset of the program:
Important molecular pathology tests were often not performed at diagnosis in cancer patients in UK. For instance, patients with newly diagnosed metastatic melanoma would often have to wait several weeks before discovering the BRAF mutation status of their cancer, and their consequent eligibility for BRAF-targeted treatment. This is a period of considerable anxiety for patients and those treating them. Approximately half of the patients are ineligible for BRAF inhibition, but will still have delays in their other treatment while awaiting the results of molecular testing. These delays would be avoidable with a nationwide infrastructure facilitating such tests at the point of diagnosis for all cancer patients.
Cancer biopsy samples were not being exploited to their full clinical or translational potential outside of clinical trials. Despite a dramatic increase in UK cancer trial recruitment during the past decades Citation[10], there was no infrastructure in existence for systematically interrogating the majority of tissue samples from patients who were not recruited to studies. A key aim of SMP was to define the core genetic characteristics of this population, correlating it with demographic, diagnostic, treatment and outcome data, which was often routinely recorded during healthcare delivery.
Progress & findings of SMP Phase I
Phase I of the programme started in 2011 and was completed in 2013. The programme was approved by the National Research Ethics Service Committee East of England (ref 11/EE/0202). Patients were identified as eligible following tissue diagnosis of one of six main tumor types (breast, colorectal, prostate, lung or ovarian cancer or malignant melanoma) while receiving treatment at one of 26 UK hospitals linked to a network of eight regional Experimental Cancer Medicine Centres. These patients gave consent to centralized molecular analysis of surplus formalin-fixed, paraffin-embedded material from their diagnostic biopsy or resection specimen. In total, 10,754 cancer patients, 98% of those approached, consented to analysis of their diagnostic material between August 2011 and July 2013, and 9010 samples were sent for analysis. Each tissue sample was forwarded with a matched peripheral blood sample from each patient to one of three technical hubs where they were analyzed for the presence of the following small panel of key genetic abnormalities determined according to tumor type:
Lung cancer: KRAS, EGFR, ALK, DDR2, BRAF.
Breast cancer: PTEN, PIK3CA, BRAF, TP53.
Ovarian cancer: PTEN, PIK3CA, BRAF, TP53.
Colorectal cancer: KRAS, NRAS, BRAF, PIK3CA, TP53.
Melanoma: BRAF, NRAS, PIK3CA, KIT.
Prostate cancer: PTEN, BRAF, TMPRSS2-ERG.
Point mutations (BRAF, DDR2, EGFR, KIT, KRAS, NRAS, PIK3CA, PTEN and TP53) were determined using Sanger sequencing, pyrosequencing or comparable methods, according to each laboratory’s experience and preference, and gene rearrangements (ALK and TMPRSS2-ERG) by fluorescent in situ hybridization. Results were then transmitted electronically to clinical centers for inclusion in patients’ medical records, and a clinical dataset including diagnostic, treatment and outcome data was collated for all patients using the newly approved English national cancer registration dataset. Analysis of the data to date shows that even with this small panel of genes, at least one genetic abnormality was found in 53% of cases. Although the limitations of performing DNA-based analysis in formalin-fixed material are widely acknowledged, only 3% of samples failed all tests.
By completing this process for a series of thousands of patients across the UK, we confirmed the feasibility of incorporating a platform for molecular diagnostics in the normal pathway of cancer care and demonstrated the applicability of a nationwide testing network with results linked to healthcare records. This task presented a number of practical difficulties, with a number of recommendations gathered from the process that could inform the development of future initiatives:
Standards for molecular pathology
Standards for sampling, tissue fixation, processing, nomenclature and reporting terminology must be established to describe genetic abnormalities present in tumors.
Consensus should be sought from experts in histopathology and molecular genetics on the circumstances in which tissue macrodissection is required prior to somatic mutation testing (e.g., when tissue tumor content is less than 30%).
Standards for validation of new technologies and assays must be agreed upon by experts in molecular genetics. There should also be mandatory ongoing participation in a programme of external quality assessment for laboratories providing this service to patients.
Test turnaround times and workload efficiency must be balanced out according to clinical need and by batching of samples where appropriate.
Educational resources about molecular pathology and the diagnostic applications of emerging genetic technologies must be supplied to histopathologists and biomedical scientists. There should be adequate representation of these areas in the relevant training curricula to equip the future workforce with the necessary knowledge and skills.
There is a need for a central regularly updated and rigorously curated online reference resource featuring detailed information on key genes implicated in cancer including genetic polymorphisms and mutations with evidence for their functional significance.
Data handling
Electronic test requesting and reporting should be used to provide a secure, acceptable and effective means of communication between different organizations, and be adopted for molecular diagnostics work.
Routine consent for research use of surplus material from diagnostic tissue samples should become routine in practice and recorded within the patient record for ease of reference.
Molecular data should be integrated with histopathological data and provided to the clinical team in a single comprehensive report to determine appropriate patient management.
The creation of a National Cancer Registration Service in England represents a unique opportunity to enable the collection of high-quality, accurate and complete data from routine clinical care, which can then be de-identified and linked to stored surplus tissue and molecular data for research use and ultimately patient benefit.
Commissioning, funding & evaluating an evidence-based service
The cost of molecular testing should be budgeted by healthcare commissioners at a national level, with sufficient funding made available to pathology departments to establish and deliver the service.
Clinical trials
To facilitate testing for a single panel of key genetic abnormalities driving different cancer types and identify trials available to patients whose tumors test positive for these genetic abnormalities, the national trials portfolio should be developed to facilitate identification of clinical trials across multiple tumor types with eligibility determined by the presence of a particular genetic abnormality (so-called ‘basket’ trials).
Next steps: SMP Phase II
This year we will commence Phase II, a project that will build on many aspects of the infrastructure and findings from Phase I. We will provide molecular prescreening of samples from patients with advanced stage non-small cell lung cancer, on a single multiplexed next-generation sequencing platform for a range of key genetic abnormalities, including gene copy number aberrations, as well as mutations and structural rearrangements. The results of this analysis will be used to inform patient eligibility for a nationwide multiarm molecularly stratified clinical trial (the National Lung Matrix Trial), opening at the end of 2014 across multiple Experimental Cancer Medicine Centre sites in the UK. To adapt to the rapidly changing landscape of genetic technology, a multiplex next-generation sequencing panel has been piloted with samples from the last 3 months of SMPI, showing high levels of concordance in comparison to conventional Sanger sequencing. The key aim of SMP2 will be to demonstrate the feasibility of combining molecular testing with clinical trial enrolment and translational progress on a national level. We believe that this approach will be critical to the progress of research in targeted molecular therapy for the future.
Acknowledgements
Funding for the Stratified Medicine Programme is acknowledged from Cancer Research UK and programme founding partners AstraZeneca and Pfizer. Assistance with hosting and handling the data by the National Cancer Registration Service Eastern Office and the University of Oxford Department of Computer Science is gratefully acknowledged.
Financial & competing interests disclosure
C Lindsay is funded by Cancer Research UK; PW Johnson is a board member of Cancer Research UK; and E Shaw and I Walker are employees of Cancer Research UK. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
References
- Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417(6892):949-54
- Flaherty KT, Puzanov I, Kim KB, et al. ‘Inhibition of mutated, activated BRAF in metastatic melanoma.’ N Engl J Med 2010;363(9):809-19
- Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001;344(14):1031-7
- Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med 2009;361(2):123-34
- Rosell R, Carcereny E, Gervais R, et al. Erlotinib versus standard chemotherapy as first-line treatment for European patients with advanced EGFR mutation-positive non-small-cell lung cancer (EURTAC): a multicentre, open-label, randomised phase 3 trial. Lancet Oncol 2012;13(3):239-46
- Janne PA, Ramalingam SS, Chih-Hsin Yang J, et al. Clinical activity of the mutant-selective EGFR inhibitor AZD9291 in patients (pts) with EGFR inhibitor–resistant non-small cell lung cancer (NSCLC). J Clin Oncol 2014;32(5 Suppl):abstr 8009
- Sequist LV, Soria JC, Gadgeel SM, et al. First-in-human evaluation of CO-1686, an irreversible, highly selective tyrosine kinase inhibitor of mutations of EGFR (activating and T790M). J Clin Oncol 2014;32(5 Suppl):abstr 8010
- Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366:883-92
- Nowak F, Soria JC, Calvo F. Tumour molecular profiling for deciding therapy-the French initiative. Nat Rev Clin Oncol 2012;9(8):479-86
- Cooper M, Poole K, Cameron DA, et al. Five-fold increase in recruitment of cancer patients to National Institute for Health Research cancer research network portfolio studies between 2001 and 2011. J Clin Oncol 2012;30(15 Suppl):abstr 6091