Keynote comment: Reimbursement for molecularly targeted anticancer agents

Reflection and Reaction

Keynote comment: Reimbursement for molecularly targeted anticancer agents

See Policy Watch page 22

Various new molecularly targeted drugs have been approved for treatment of cancer and other nonmalignant diseases in Europe and the USA, and some have shown impressive results for response rate, progressionfree survival, disease-free survival, and overall survival even in patients with metastatic disease. In particular, patients with tumours that express the targets for these treatments (eg, the ERBB2 receptor in some breast cancers, KIT mutations in gastrointestinal stromal tumours, and the BCR–ABL1 fusion gene in patients with chronic myeloid leukaemia) are especially sensitive. However, how to expand access of these agents to more patients in a cost effective way is a matter of debate. Patients and the scientific and medical communities have pressurised the health-care system to increase the speed at which these highly active agents are approved for reimbursement. In fact, the UK National Institute for Health and Clinical Excellence is considering the introduction of a fast-track scheme for cancer drugs that improve survival. However, although regulatory agencies need to be responsive to public concerns, they should also be be free from media influence, or political, financial, or other special interests (eg, personal interests). Independence and scientific rigour are, and remain, essential. Regulatory agencies will also need to be prepared to assess a large variety of new biological agents

Global negotiations needed to tackle issues with drug reimbursement


as a consequence of advances in molecular biology and clinical drug development, and health-care systems will need to have a long-term strategy (ie, 10–20 years) to implement these changes, rather than short-term strategies that are often based on political factors. The high costs of these treatments are a substantial obstacle to their widespread use, posing challenges to clinicians who care for patients with cancer. For instance, the cost of 8 weeks’ treatment for metastatic colorectal cancer with FOLFOX (folinic acid, fluorouracil, and oxaliplatin) and bevacizumab given every 2 weeks can reach US$21 033 in the USA.1 To overcome this serious issue, identification of patients who are most likely to benefit from these treatments should be the primary objective of academic researchers and the pharmaceutical industry, rather than a secondary objective. For example, although gefitinib has limited activity in non-small-cell lung cancer, striking responses have been seen in patients who have tumours with specific activating mutations in the tyrosine-kinase domain of the epidermal-growthfactor receptor.2 Making identification of such patients a priority should lead to less empirical and more individualised prescriptions as soon as possible. The high costs of these agents should be reconsidered seriously and renegotiated regularly. For example, as the number of patients given these drugs increases, the cost of treatment should decrease, ensuring better access to the new therapies for all patients in poor and rich countries, and economic safety for the pharmaceutical industry because new drug development is a long, hard, and costly process. We also believe that the time covered by specific patents should be redressed for at least some of these agents. Some governmental initiatives on patent exemptions have been successful and could be extended to molecularly targeted treatments—for example, all drugs for HIV/AIDS and their manufacturing processes have been exempt from patent coverage in Brazil since June, 2005.3 Global negotiations, such as those implemented by the European Medicines Evaluation Agency network, will probably be more promising and applicable than are those undertaken by individual groups. We suggest that rigorous and realistic pharmacoeconomic studies must be part of the new drug Vol 7 January 2006

Reflection and Reaction

development and marketing process: efficacy, safety, and cost-effectiveness of new drugs must be periodically assessed.4 In this way, we could decrease drug-related costs and increase the number of patients treated.

Clinic, Jules Bordet Institute, Brussels, Belgium (AA) [email protected] We declare no conflicts of interest. 1 2

Umberto Tirelli, *Gilberto de Castro Jr, Ahmad Awada Department of Medical Oncology, National Cancer Institute, Aviano, Italy (UT); Clinical Oncology Service, Radiology Department—InRad, Hospital das Clinicas, University of São Paulo Medical School, Avenida Doutor Eneas de Carvalho Aguiar, 255, 3rd floor, room 7.21, São Paulo, Brazil 05403-001 (GC); and Medical Oncology

3 4

Schrag D. The price tag on progress: chemotherapy for colorectal cancer. N Engl J Med 2004; 351: 317–19. Lynch TJ, Bell DW, Sordella R. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer. N Engl J Med 2004; 350: 2129–39. Ahmad K. Brazil takes a step towards patent exemption for HIV drugs. Lancet Infect Dis 2005; 5: 399. Neyt M, Albrecht J, Cocquyt V. An economic evaluation of Herceptin in adjuvant setting: the Breast Cancer International Research Group 006 trial. Ann Oncol 2005 (published online Nov 30, 2005; DOI: 10.1093/annonc/mdj101).

Inherited cytokine response and risk of lymphoma In this issue of The Lancet Oncology,1 a report by the InterLymph Consortium addresses the role of genetic determinants of the immune response in the risk of developing non-Hodgkin lymphoma. The researchers found that the risk of diffuse large-B-cell lymphoma was significantly increased in individuals with tumour necrosis factor (TNF) 308G→A and the interleukin 10 (IL10) 3575T→A polymorphisms. Given the numbers of cases (n=3586) and the methods used, this study clearly overcomes the limitations of smaller non-populationbased studies that assessed the distribution of such singlenucleotide polymorphisms in patients with lymphoma. Two points, however, should be mentioned. First, this study included only white patients—a precaution outlined in the methods as a result of the geographic variability of genotype distribution. Therefore, the findings cannot be applied to populations of other ethnic origins and might be diluted eventually in countries with substantial population mixing. Second, the effects of the TNF and IL10 polymorphisms were recorded mainly for diffuse large-B-cell lymphoma, and TNF 308G→A was not associated with increased risk of follicular lymphoma. The low frequency of other lymphoma subtypes and difficulties of diagnostic reproducibility in the absence of pathological review by a panel prevented the researchers from estimating risk for these subtypes. However, assessment of the role of genetic determinants of the immune system in lymphoma subtypes such as mucosaassociated lymphoid tissue and marginal-zone B-cell lymphomas, which might be triggered by microbial infections, will be of interest. At present, no preventive action can be taken to limit the risk of lymphoma, and early screening does not exist Vol 7 January 2006

for this disease. Therefore, these findings cannot justify any attempt to identify individuals at risk of lymphoma in general practice; however, they give some important insights into the pathogenesis of lymphoma. Raised concentrations of TNF and IL10 confer a poor outlook for patients with lymphoma, and the genetic background regulating these cytokines is associated with patient outcome. However, the mechanism by which the genetic background controlling TNF and IL10 production could affect the onset of lymphoma remains to be elucidated. Moreover, how these polymorphisms change cytokine production is not clear. The TNF 308G→A polymorphism is associated with increased TNF production and might affect gene transcription directly,2–4 although some data challenge these observations.5 Other genes located within the HLA type III region are subject to variations in linkage disequilibrium with TNF 308G→A and might also play a part in risk of non-Hodgkin lymphoma. Despite the known genetic inheritance for IL10 production, its relation with IL10 haplotypes has not been characterised fully, and the alleles at nucleotide 3575 alone cannot account such inherited variability.6 The association between haplotypes containing TNF 308G→A and IL10 3575T→A polymorphisms, with a phenotype characterised by high TNF production and low IL10 production or an unbalanced T-helper (Th)1/Th2 (ie, cellular vs humoral immunity) response, still needs to be confirmed. Furthermore, we need to understand how these haplotypes and their phenotype could explain an increased risk of lymphoma. One idea is that a biased Th1 inflammatory process over than of a Th2 reaction augments the immune response against some antigens or pathogens. How this increased immune response might

See Articles page 27