original articles
Annals of Oncology Annals of Oncology 25: 270–276, 2014 doi:10.1093/annonc/mdt514
Selection of oncology medicines in low- and middle-income countries Y. T. Bazargani1, A. de Boer1, J. H. M. Schellens1,2, H. G. M. Leufkens1 & A. K. Mantel-Teeuwisse1* 1 Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht; 2Division of Clinical Pharmacology, The Netherlands Cancer Institute, Amsterdam, The Netherlands
Background: High cancer mortality rates in low- and middle-income countries (LMICs) have raised concerns regarding access to oncology medicines. Essential medicines are those which satisfy the primary health care needs and provide a basis for public procurement or reimbursement decisions in LMICs. We explored selection of oncology medicines in LMICs through investigating national essential medicines lists (NEMLs) for cancer treatments. Methods: Recently updated NEMLs were retrieved for 76 countries. Oncology medicines were classified based on therapeutic categories. Countries were clustered based on geographic regions, income levels and burden of cancer (mortality and morbidity). Indicators of frequency (number of medicines), diversity (number of therapeutic (sub)categories) and more importantly absence were measured and compared across countries using parametric and nonparametric tests. Results: The overall median number of oncology medicines on NEMLs was 16 (interquartile range = 23) chosen predominantly from subcategories of ‘antineoplastic agents’, with substantial variation across regions and income groups. Five countries did not select any oncology medicine and 68% did not have any ‘hormones and related agents’ on their NEMLs. Newer technologies like targeted therapies were infrequently incorporated. The cluster of countries suffering most from the burden of cancer selected more essential oncology medicines and diversified further. Conclusion: The observed selection of oncology essential medicines can reflect insufficiencies and inequalities in access to cancer treatments at least in the public sector of LMICs. Further resources need to be allocated from governments and international organizations to tackle the problem of access to oncology medicines in these countries. Key words: oncology medicines, low and middle income countries, selection, essential medicines, access to medicines
introduction Health care systems have incurred a massive burden from cancer. Currently, cancer is the second cause of death, accounting for 13% of all casualties worldwide [1]. The global cost of newly diagnosed cancer cases was estimated to exceed $US 300 billion in 2010 [1]. Enormous investments have been made in research and development for oncology medicines. For instance, a great number of projects worth over $US 5 billion were funded by the US National Cancer Institute (NCI) in this respect in 2011 [2]. Despite substantial endeavors in the field, little is known about the global situation of access to oncology medicines particularly in developing and deprived countries. 70% of cancer mortality occurs in low- and middle-income countries (LMICs) [3]. This is undoubtedly attributable to socioeconomic determinants which may cause issues like lack of screening and routine examinations and subsequently late-stage cancer at the time of *Correspondence to: Dr Aukje K. Mantel-Teeuwisse, Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, PO Box 80082, 3508 TB Utrecht, The Netherlands. Tel: +31-30-2537324; Fax: +31-30-253-9166; E-mail: [email protected]
diagnosis [4]. However, sparse number of studies conducted in some LMICs has unanimously addressed the lack of access to medicines as an outstanding issue [5, 6]. A comprehensive national cancer program which encompasses different aspects of prevention, detection and treatment has been suggested as a sustainable solution [7]. Proper adaptation of such plans—which could ultimately ensure treatment benefits supported by evidence—is crucial but would impose a catastrophic additional expenditure on the resource constrained health care systems which are simultaneously confronted with widespread communicable disease and undernutrition [8]. Essential medicines are supposed to ensure access to medicines needed for priority health care issues at a national (or subnational) level in LMICs. National essential medicines lists (NEMLs) are considered the basis for public procurement or reimbursement purposes. Particularly, in the poorer countries, the list plays a pivotal role in terms of prioritizing public expenditures on medicines. According to the World Health Organization (WHO) over 90% of surveyed low- and middleincome countries used their NEML for public procurement [9]. Besides, essential medicines are provided by public procurement or social health insurance schemes free of charge in two-third of
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
Received 16 August 2013; revised 23 September 2013; accepted 24 September 2013
original articles
Annals of Oncology
methods data collection and classification The last available update of each NEML was obtained from the ‘WHO database of essential medicine lists and formularies’ [12]. WHO has recommended countries to update their NEMLs periodically, hence NEMLs not updated since 2005 were excluded to ensure that only dynamic lists were considered [9]. In China, provincial EMLs were combined to make an EML list, in the absence of essential oncology medicines on the NEML of China [13]. Eventually, 76 countries were included in the analysis (supplementary Appendix 1, available at Annals of Oncology online). Medicines were included in the study if they were classified as oncology medicines in the NEML (or equivalent terms in different NEMLs or languages). Palliative and supportive therapies and medicines for management of side-effects and complications were excluded. The medicines were classified according to the Anatomical Therapeutic Chemical (ATC) classification for further analysis [14]. Oncology medicines fall predominantly under the ATC class L ‘antineoplastic and immunomodulating agents’, namely L01 Antineoplastic agents; L02 Endocrine therapy; L03 Immunostimulants and L04 Immunosuppressants (supplementary Appendix 2, available at Annals of Oncology online). A category ‘others’ was created for the remaining medicines assigned as oncology medicines. The burden of cancer (mortality and morbidity) was obtained for each country from the WHO global burden of disease database. Casualty figures (2008) and disability-adjusted life years (DALYs, 2004) were estimated by cause for each WHO Member State [15]. Data on geographic regions and income levels were obtained from WHO and the World Bank, respectively [16, 17].
data analysis The total number of essential oncology medicines and ATC (sub)categories in the studied countries was obtained and analyzed according to the aforementioned classifications. Countries were stratified by WHO geographic region, the latest World Bank income group classification and relative burden of cancer for further analysis. Relative burden of cancer was estimated as a proportion of overall burden of disease in each country, and countries were then grouped almost equally according to cancer mortality and morbidity. Where the numbers of medicines were compared in different clusters of countries, parametric (ANOVA test for normally distributed data) or nonparametric tests (Kruskal–Wallis test for non-normal distributions) were used to investigate the difference among various groups. Regression analysis (linear and log-linear regression) was employed to examine the association between cancer burden and the number of listed oncology essential
Volume 25 | No. 1 | January 2014
medicines within each country. All statistical analyses were conducted using SPSS software, version 19.
results Overall, the median number of oncology medicines listed in the NEMLs of the studied countries was 16 (range 0–49; interquartile range (IQR) 23, see Table 1). The median number of ATC categories from which the medicines were selected was 2 (range: 0–4). Five countries (7%) did not have any essential oncology medicine on their NEML. ‘Antineoplastic agents’ were most widely included (median: 15, range 0–41), while ‘endocrine therapy’ (median: 1, range: 0–10) and especially ‘immunostimulants’ and ‘immunosuppressants’ were poorly represented. Over 75% of the countries did not have any oncology medicines in the two latter mentioned categories, neither any ‘taxanes’, ‘monoclonal antibodies’ and ‘protein kinase inhibitors’. More detailed information can be found in supplementary Appendices 3 and 4, available at Annals of Oncology online. The number of oncology essential medicines differed significantly between geographical regions. The median (IQR) number of essential oncology medicines was highest in the region of the Americas with 30 [18] and lowest in the western Pacific region with just 3.5 (17, P < 0,001). Similar patterns were observed for different subgroups of oncology medicines (Table 1). More than 70% of the countries in all regions selected medicines from two or more of the ATC categories (except 54% of the African countries). Figure 1A shows main absence of subcategories in different geographic regions. The Western Pacific, eastern Mediterranean and African regions most frequently excluded various subcategories. The two latter regions had the highest number of countries without any endocrine therapy (46% and 30%, respectively). While 12 (80%) of the countries in the region of the Americas had at least one taxane on the list, over 80% in the African (n = 21) and over 90% in the Western Pacific (n = 11) regions had not included any. Monoclonal antibodies were also not included in >50% of the studied European and American countries, as well as over 75% of the countries in all other regions. Similarly, >75% of the countries in all regions (except Europe) did not have any protein kinase inhibitors listed as essential medicine. The median (IQR) number of essential oncology medicines differed between low- (n = 11 [13]), lower-middle (n = 18 [19]) and upper-middle income (n = 26 [34]) countries, although statistical significance was just not reached (Table 1). Trends were similar for antineoplastic agents and endocrine therapy. At least 75% of the low-income and half of the middle-income countries did not designate any taxanes as essential oncology medicines. No difference was observed across income levels when inclusion of monoclonal antibodies, protein kinase inhibitors and ‘α-interferons’ was considered. Notable absent subgroups in different income levels are given in Figure 1B. Endocrine therapy was not selected by 35% of low-income countries. For ‘hormones and related agents’, observed absence was 85% in low-income countries versus 61% in middle-income countries. Countries with >12% cancer mortality had more listed oncology essential medicines than countries with ‘20% 100
“Plant alkaloids and other natural products”
Percentage of absence
90 80 70
“Cytotoxic antibiotics and related compounds”
60 50
“Other antine oplastic agents”
40
“Hormones and related agents”
30
“Hormone antagonists and related agents”
20 10 0 Low income
Lower middle incomeUpper middle income
Figure 1. (A) Absence of subcategories of oncology medicines on NEMLs in WHO regions. (B) Absence of subcategories of oncology medicines on NEMLs in income levels.
Table 2. Oncology essential medicines for different disease burden groups of countries
Cancer mortality as percentage of total mortality Cancer morbidity as percentage of total morbidity
Country groups
Number of countries
Mean oncology essential medicines on the list
SD
P value (ANOVA)
P value (ANOVA)
12% 10%
25 30 21 28 25 23
12 16.5 27 12.3 15.7 27.2
11.8 12.6 14.8 9.5 12.9 16
reference 0.447 0.001 reference 0.623 0.001
– reference 0.022 – reference 0.028
were similar for cancer morbidity measures as well as subcategories (antineoplastic agents and endocrine therapy) of medicines (Table 2). Figure 2A and B shows the association between the number of essential oncology medicines and mortality and morbidity fractions due to cancer. Regression coefficients were 0.78 and 2.52, respectively (P < 0.001 and P = 0.007, respectively), which means an increase in cancer mortality or morbidity fraction by
Volume 25 | No. 1 | January 2014
10% is associated on average with an addition of 8 or 25 oncology medicines on the corresponding NEML, respectively. A total of 52%, 80% and 95% of the countries with 12% mortality due to cancer have chosen their medicines from two or more ATC categories. A similar trend was observed in the clusters of countries grouped by cancer morbidity, where the respective figures were 54%, 84% and 91% of the countries.
doi:10.1093/annonc/mdt514 |
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
Af
er
ic
ric
a
n
“Other antineoplastic agents” a
Percentage of absence
A
original articles
Annals of Oncology
number of oncology essential medicines
A 60 50 40 30 20 10
2%
4%
6%
8%
10%
12%
14%
morbidity (DALYs) fraction (cancer morbidity/all causes morbidity)
number of oncology essential medicines
B 60 50 40 30 20 10 0 0%
5%
10%
15%
20%
25%
30%
35%
40%
moratlity fraction (cancer mortality/all causes mortality) Figure 2. (A) Cancer morbidity (DALYs) fraction versus number of essential medicines per country. (B) Cancer mortality fraction versus number of essential medicines per country.
discussion Overall, the median number of selected essential oncology medicines was 16, with NEMLs being dominated by antineoplastic agents. Five countries did not select any oncology medicine. The overall number and diversity of essential oncology medicines increased by improvement of the income level of the countries and increase in cancer burden and varied across regions. Surprisingly, subcategories of antineoplastic agents frequently used as routine chemotherapy agents were not found among essential medicines in a notable fraction of the studied countries. This category includes medicines which have extensive indications in various types of malignancies. The presented results should not be interpreted solitarily. Cancer therapy might be provided in different settings in LMICs and might not be restricted to (essential) medicines. However, the sustainability of such provisions needs to be scrutinized. The majority of the LMICs did not have any monoclonal antibodies, protein kinase inhibitors or ‘α-interferon’ on their NEMLs, neither had over half of them any taxanes which may imply a lag in adopting newer therapies. This might have changed recently by introduction of generic taxanes and may be improved further by invasion of biosimilars in the near future. Carefully conducted comparative studies are required for adopting such therapies in LMICs considering all controversies including the added clinical value and their cost-effectiveness [20].
| Bazargani et al.
Besides, acquiring new therapies without paying adequate attention to its requirements, might not utilize resources in the most efficient manner [21]. Personalized medicine offers considerable benefits to patients and health care systems by distinguishing between individual responses to a treatment in terms of effectiveness and safety. HER-2 expression and KRAS mutation tests which have high predictive values for trastuzumab and cetuximab therapies are two examples among others [22]. It is worth investigating if LMICs which encompassed new therapies have considered prerequisites of such therapies in their guidelines and their routine clinical practice. Absence of endocrine therapy was observed more in lowincome countries which seems worrisome as these medicines are extensively used in certain types of breast and prostate cancer [18, 19]. Even in the region of the Americas where according to our study hormone antagonists are well presented in NEMLs, empirical evidence has indicated lack of access to aromatase inhibitors—the first choice of endocrine therapy in hormone receptor positive breast cancer particularly in postmenopausal women—in the public sector of upper middle-income countries [23, 24]. Prostate cancer, however, is most abundant in developed countries [25]. LMICs have probably prioritized among different cancer types which might explain the absence of gonadotropin-releasing hormone agonists on many NEMLs. When countries were grouped according to cancer burden a trend was observed in which countries suffering more from
Volume 25 | No. 1 | January 2014
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
0 0%
original articles
Annals of Oncology
Volume 25 | No. 1 | January 2014
treatment of breast cancer. This is believed to contribute to the observed trend in reduction in the mortality rates [23]. On this note, we conclude that allocation of resources should be equally directed towards medicines as well as other required medical technologies and interventions, infrastructures and skilled medical staff (e.g. trained oncologists). In conclusion, this study provides evidence that the selection of oncology medicines on NEMLs is suboptimal in LMICs which consequently may impact access to cancer treatments. This finding does not only apply to newer medicines, it also includes routine chemotherapies and in particular hormone therapies. Countries in which cancer is more burdensome have included more oncology medicines and diversified their NEMLs further compared with other countries. Adequate attention from the national health care decisionmakers and international organizations is deemed necessary to provide a sufficient amount of oncology medicines to LMICs in hope to reduce death toll due to cancer in an equitable manner.
disclosure The authors have declared no conflicts of interest.
references 1. Bloom DE, Cafiero ET, Jané-Llopis E et al. The Global Economic Burden of Noncommunicable Diseases. Geneva: World Economic Forum 2011 (www. weforum.org/EconomicsOfNCD). 2. National Cancer Institute. NCI funded research portfolio; http://fundedresearch. cancer.gov/search/funded?Action=full&fy=PUB2011&type=sic (18 November 2012, date last accessed). 3. World Health Organization. Cancer fact sheet. 2013; http://www.who.int/ mediacentre/factsheets/fs297/en/ (10 August 2013, date last accessed). 4. Clegg LX, Reichman ME, Miller BA et al. Impact of socioeconomic status on cancer incidence and stage at diagnosis: selected findings from the surveillance, epidemiology, and end results: National Longitudinal Mortality Study. Cancer Causes Control 2009; 20(4): 417–435. 5. Denny L. Cervical cancer treatment in Africa. Curr Opin Oncol 2011; 23(5): 469–474. 6. Zucca E, Rohatiner A, Magrath I et al. Epidemiology and management of lymphoma in low-income countries. Hematol Oncol 2011; 29(1): 1–4. 7. World Health Organization. National cancer control programmes. 2013; http ://www.who.int/cancer/nccp/en/ (25 July 2013, date last accessed). 8. Bygbjerg IC. Double burden of noncommunicable and infectious diseases in developing countries. Science 2012; 337(6101): 1499–1501. 9. van den Ham R, Bero L, Laing R. The World Medicines Situation 2011; selection of essential medicines; http://apps.who.int/medicinedocs/en/m/abstract/ Js18770en/ (4 December 2012, date last accessed). 10. World health Organization. Country profiles and monitoring of the pharmaceutical situation in countries; http://www.who.int/medicines/areas/coordination/ coordination_assessment/en/index.html (23 November 2012, date last accessed). 11. World Health Organization. Comparative table of medicines on the WHO Essential Medicines Listfrom 1977–2011; http://www.who.int/selection_medicines/list/en/ (4 December 2012, date last accessed). 12. World health Organization. Essential Medicines Lists and Formularies; http://www. who.int/selection_medicines/country_lists/en/index.html (23 November 2012, date last accessed). 13. Wang L, Ma E, Xu W. Comparative Analysis of China National & Twenty-two Selected Provincial Essential Medicine Lists to the WHO 2011 Model List; http ://apps.who.int/medicinedocs/en/m/abstract/Js18851en/ (23 November 2012, date last accessed). 14. WHO collaborating center for Drug Statistics Methodology. ATC/DDD Index 2012; http://www.whocc.no/atc_ddd_index/ (23 November 2012, date last accessed).
doi:10.1093/annonc/mdt514 |
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
cancer burden selected more essential oncology medicines and diversified their choices from different categories. This might indicate that decision making procedures are reasonably informed about the national health care priorities. However, our regression model suggests that variation in selecting oncology essential medicines can only to a limited extent be explained by variation in cancer burden (very low R 2), thus further studies are required to assess which selection decisions are influenced by disease burden on a national scale. One might even suggest that an overall package of anticancer medicines of a country can be reasonably well balanced despite all the exclusions, taking national cancer incidence patterns into account [26]. There were some limitations in our study. First, medicines complementary to chemotherapy or used for supportive and palliative care were beyond the scope of our study. Owing to the late presentation of cancer in LMICs, palliative care medicines get further priority and should be carefully positioned in a thorough overview. Secondly, we studied NEMLs at a national level, but oncology medicines might also be selected at a subnational level, e.g. in hospital formularies or therapeutic guidelines. Another limitation was the inability to measure actual availability of oncology medicines. Direct measurement of availability and affordability of oncology medicines has not been studied so far in LMICs. Perhaps lack of information and transparency particularly in the oncology field where both inpatient and outpatient data are required explains the absence of such data. Moreover, due to constant lack of health care resources in the public sector, one might argue that a situation analysis in the private sector is also crucial to provide a comprehensive insight. Substantial disparities in access and quality of treatments have been reported in both targeted therapies as well as chemotherapies among the two sectors [24]. Besides, due to massive discrepancies in socioeconomic status of different walks of life in many LMICs, access to medicines and care is not consistent within a country, which makes it difficult to draw an integrated conclusion even at a national level [27]. It is arguable that cancer morbidity and mortality data used in this analysis might not reflect the actual situation due to lack of information, lack of patient registries and underdiagnoses in LMICs. Nevertheless, the WHO data on burden of disease are the most extensive available data which have been reliably used in similar global and regional studies [28]. Lastly, medicines are just one component of care. Treatment of cancer is a multifactorial issue and requires an extensive network of screening, diagnosis, monitoring and different treatment strategies (e.g. surgery, radiotherapy) alongside medicines. There is no evidence to suggest that these components—which are even more resource intensive—are convincingly available in LMICs in an equitable manner [5, 6, 29]. One might argue that constrained health care resources should be allocated with a comprehensive approach to fulfill all the components of cancer treatment. Despite such a seemingly reasonable universal recipe for all types of cancers, there are some contrary evidence-based opinions in case of breast cancer suggesting superiority of treatment over screening in resource limited countries [30, 31]. Similar arguments might be applicable to different situations and cancer types and hence worthwhile investigating further. Mexico can be regarded as a learning case, where health policy reform has led to improved access to screening, diagnosis and
original articles
23. Chavarri-Guerra Y, Villarreal-Garza C, Liedke PE et al. Breast cancer in Mexico: a growing challenge to health and the health system. Lancet Oncol 2012; 13(8): e335–e343. 24. Lee BL, Liedke PE, Barrios CH et al. Breast cancer in Brazil: present status and future goals. Lancet Oncol 2012; 13(3): e95–e102. 25. American Cancer Society. Global Cancer Facts & Figures. 2011; http://www. cancer.org/acs/groups/content/@epidemiologysurveilance/documents/document/ acspc-027766.pdf (20 December 2012, date last accessed). 26. Sikora K, Advani S, Koroltchouk V et al. Essential drugs for cancer therapy: a World Health Organization consultation. Ann Oncol 1999; 10(4): 385–390. 27. Pedraza AM, Pollan M, Pastor-Barriuso R et al. Disparities in breast cancer mortality trends in a middle income country. Breast Cancer Res Treat 2012; 134 (3): 1199–1207. 28. Patel AR, Prasad SM, Shih YC et al. The association of the human development index with global kidney cancer incidence and mortality. J Urol 2012; 187(6): 1978–1983. 29. Agarwal G, Ramakant P, Forgach ER et al. Breast cancer care in developing countries. World J Surg 2009; 33(10): 2069–2076. 30. Autier P, Boniol M, Gavin A et al. Breast cancer mortality in neighbouring European countries with different levels of screening but similar access to treatment: trend analysis of WHO mortality database. BMJ 2011; 343: d4411. 31. Gøtzsche PC, Nielsen M. Screening for breast cancer with mammography. Cochrane Database Syst Rev 2011; (1): CD001877. DOI:10.1002/14651858. CD001877.pub4.
Annals of Oncology 25: 276–282, 2014 doi:10.1093/annonc/mdt524
The prognostic significance of left ventricular ejection fraction in patients with advanced cancer treated in phase I clinical trials R. Said1, J. Banchs2, J. Wheler1, K. R. Hess3, G. Falchook1, S. Fu1, A. Naing1, D. Hong1, S. Piha-Paul1, Y. Ye1, E. Yeh2, R. A. Wolff1 & A. M. Tsimberidou1* Departments of 1Investigational Cancer Therapeutics; 2Cardiology; 3Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, USA
Received 16 January 2013; revised 5 September 2013; accepted 16 October 2013
Background: New targeted agents may cause acute cardiac events. The purpose of our study was to investigate the incidence and the prognostic significance of left ventricular ejection fraction (LVEF) in phase I trials. Patients and methods: Between October 2008 and September 2011, the records of 1166 consecutive patients with advanced cancer treated in the Phase I Clinic who underwent echocardiography were retrospectively reviewed. Results: Most of the patients were White (78%), and the most common tumor types were colorectal cancer and melanoma. Of 1166 patients, 177 (15.2%) patients had an LVEF of 35%, ≤2 prior systemic therapies, ≤2 metastatic sites, and normal lactate dehydrogenase and albumin levels. Conclusion: Echocardiography would improve patient selection for enrollment in phase I clinical trials. These data suggest that it is safe to treat patients with LVEF between 35% and 50%. Key words: biomarker, cardiac dysfunction, malignancy and phase I trial *Correspondence to: Dr A. M. Tsimberidou, Department of Investigational Cancer Therapeutics, The University of Texas MD Anderson Cancer Center, Unit 455, 1515 Holcombe Boulevard, Houston, TX 77030, USA. Tel: +1-713-792-4259; Fax: +1-713794-3249; E-mail: [email protected]
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
15. World health Organization. Health Statistics and Health Information Systems; Disease and injury country estimates; http://www.who.int/healthinfo/global_ burden_disease/estimates_country/en/index.html (23 November 2012, date last accessed). 16. World health Organization. WHO geographic regions; http://www.who.int/about/ regions/en/index.html (23 November 2012, date last accessed). 17. The World Bank. World Bank classification of countries; http://data.worldbank.org/ indicator/NY.GNP.PCAP.CD (23 November 2012, date last accessed). 18. National Cancer Institute. Breast cancer treatment; treatment options overview; http://www.cancer.gov/cancertopics/pdq/treatment/breast/Patient/page5#Keypoint 24 (5 December 2012, date last accessed). 19. National Cancer Institute. Prostate cancer treatment; treatment options overview; http://www.cancer.gov/cancertopics/pdq/treatment/prostate/Patient/ page4 (5 December 2012, date last accessed). 20. Greenberg D, Earle C, Fang CH et al. When is cancer care cost-effective? A systematic overview of cost-utility analyses in oncology. J Natl Cancer Inst 2010; 102(2): 82–88. 21. Stolk P, Willemen MJ, Leufkens HG. Rare essentials: drugs for rare diseases as essential medicines. Bull World Health Organ 2006; 84(9): 745–751. 22. Blair E. Predictive tests and personalised medicine. Drug Discov World 2009; (Fall): 27–31. http://integratedmedicines.co.uk/images/presentations/E_Blair_ Predictive_tests.pdf.
Annals of Oncology
Annals of Oncology Annals of Oncology 25: 270–276, 2014 doi:10.1093/annonc/mdt514
Selection of oncology medicines in low- and middle-income countries Y. T. Bazargani1, A. de Boer1, J. H. M. Schellens1,2, H. G. M. Leufkens1 & A. K. Mantel-Teeuwisse1* 1 Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht; 2Division of Clinical Pharmacology, The Netherlands Cancer Institute, Amsterdam, The Netherlands
Background: High cancer mortality rates in low- and middle-income countries (LMICs) have raised concerns regarding access to oncology medicines. Essential medicines are those which satisfy the primary health care needs and provide a basis for public procurement or reimbursement decisions in LMICs. We explored selection of oncology medicines in LMICs through investigating national essential medicines lists (NEMLs) for cancer treatments. Methods: Recently updated NEMLs were retrieved for 76 countries. Oncology medicines were classified based on therapeutic categories. Countries were clustered based on geographic regions, income levels and burden of cancer (mortality and morbidity). Indicators of frequency (number of medicines), diversity (number of therapeutic (sub)categories) and more importantly absence were measured and compared across countries using parametric and nonparametric tests. Results: The overall median number of oncology medicines on NEMLs was 16 (interquartile range = 23) chosen predominantly from subcategories of ‘antineoplastic agents’, with substantial variation across regions and income groups. Five countries did not select any oncology medicine and 68% did not have any ‘hormones and related agents’ on their NEMLs. Newer technologies like targeted therapies were infrequently incorporated. The cluster of countries suffering most from the burden of cancer selected more essential oncology medicines and diversified further. Conclusion: The observed selection of oncology essential medicines can reflect insufficiencies and inequalities in access to cancer treatments at least in the public sector of LMICs. Further resources need to be allocated from governments and international organizations to tackle the problem of access to oncology medicines in these countries. Key words: oncology medicines, low and middle income countries, selection, essential medicines, access to medicines
introduction Health care systems have incurred a massive burden from cancer. Currently, cancer is the second cause of death, accounting for 13% of all casualties worldwide [1]. The global cost of newly diagnosed cancer cases was estimated to exceed $US 300 billion in 2010 [1]. Enormous investments have been made in research and development for oncology medicines. For instance, a great number of projects worth over $US 5 billion were funded by the US National Cancer Institute (NCI) in this respect in 2011 [2]. Despite substantial endeavors in the field, little is known about the global situation of access to oncology medicines particularly in developing and deprived countries. 70% of cancer mortality occurs in low- and middle-income countries (LMICs) [3]. This is undoubtedly attributable to socioeconomic determinants which may cause issues like lack of screening and routine examinations and subsequently late-stage cancer at the time of *Correspondence to: Dr Aukje K. Mantel-Teeuwisse, Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, PO Box 80082, 3508 TB Utrecht, The Netherlands. Tel: +31-30-2537324; Fax: +31-30-253-9166; E-mail: [email protected]
diagnosis [4]. However, sparse number of studies conducted in some LMICs has unanimously addressed the lack of access to medicines as an outstanding issue [5, 6]. A comprehensive national cancer program which encompasses different aspects of prevention, detection and treatment has been suggested as a sustainable solution [7]. Proper adaptation of such plans—which could ultimately ensure treatment benefits supported by evidence—is crucial but would impose a catastrophic additional expenditure on the resource constrained health care systems which are simultaneously confronted with widespread communicable disease and undernutrition [8]. Essential medicines are supposed to ensure access to medicines needed for priority health care issues at a national (or subnational) level in LMICs. National essential medicines lists (NEMLs) are considered the basis for public procurement or reimbursement purposes. Particularly, in the poorer countries, the list plays a pivotal role in terms of prioritizing public expenditures on medicines. According to the World Health Organization (WHO) over 90% of surveyed low- and middleincome countries used their NEML for public procurement [9]. Besides, essential medicines are provided by public procurement or social health insurance schemes free of charge in two-third of
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
Received 16 August 2013; revised 23 September 2013; accepted 24 September 2013
original articles
Annals of Oncology
methods data collection and classification The last available update of each NEML was obtained from the ‘WHO database of essential medicine lists and formularies’ [12]. WHO has recommended countries to update their NEMLs periodically, hence NEMLs not updated since 2005 were excluded to ensure that only dynamic lists were considered [9]. In China, provincial EMLs were combined to make an EML list, in the absence of essential oncology medicines on the NEML of China [13]. Eventually, 76 countries were included in the analysis (supplementary Appendix 1, available at Annals of Oncology online). Medicines were included in the study if they were classified as oncology medicines in the NEML (or equivalent terms in different NEMLs or languages). Palliative and supportive therapies and medicines for management of side-effects and complications were excluded. The medicines were classified according to the Anatomical Therapeutic Chemical (ATC) classification for further analysis [14]. Oncology medicines fall predominantly under the ATC class L ‘antineoplastic and immunomodulating agents’, namely L01 Antineoplastic agents; L02 Endocrine therapy; L03 Immunostimulants and L04 Immunosuppressants (supplementary Appendix 2, available at Annals of Oncology online). A category ‘others’ was created for the remaining medicines assigned as oncology medicines. The burden of cancer (mortality and morbidity) was obtained for each country from the WHO global burden of disease database. Casualty figures (2008) and disability-adjusted life years (DALYs, 2004) were estimated by cause for each WHO Member State [15]. Data on geographic regions and income levels were obtained from WHO and the World Bank, respectively [16, 17].
data analysis The total number of essential oncology medicines and ATC (sub)categories in the studied countries was obtained and analyzed according to the aforementioned classifications. Countries were stratified by WHO geographic region, the latest World Bank income group classification and relative burden of cancer for further analysis. Relative burden of cancer was estimated as a proportion of overall burden of disease in each country, and countries were then grouped almost equally according to cancer mortality and morbidity. Where the numbers of medicines were compared in different clusters of countries, parametric (ANOVA test for normally distributed data) or nonparametric tests (Kruskal–Wallis test for non-normal distributions) were used to investigate the difference among various groups. Regression analysis (linear and log-linear regression) was employed to examine the association between cancer burden and the number of listed oncology essential
Volume 25 | No. 1 | January 2014
medicines within each country. All statistical analyses were conducted using SPSS software, version 19.
results Overall, the median number of oncology medicines listed in the NEMLs of the studied countries was 16 (range 0–49; interquartile range (IQR) 23, see Table 1). The median number of ATC categories from which the medicines were selected was 2 (range: 0–4). Five countries (7%) did not have any essential oncology medicine on their NEML. ‘Antineoplastic agents’ were most widely included (median: 15, range 0–41), while ‘endocrine therapy’ (median: 1, range: 0–10) and especially ‘immunostimulants’ and ‘immunosuppressants’ were poorly represented. Over 75% of the countries did not have any oncology medicines in the two latter mentioned categories, neither any ‘taxanes’, ‘monoclonal antibodies’ and ‘protein kinase inhibitors’. More detailed information can be found in supplementary Appendices 3 and 4, available at Annals of Oncology online. The number of oncology essential medicines differed significantly between geographical regions. The median (IQR) number of essential oncology medicines was highest in the region of the Americas with 30 [18] and lowest in the western Pacific region with just 3.5 (17, P < 0,001). Similar patterns were observed for different subgroups of oncology medicines (Table 1). More than 70% of the countries in all regions selected medicines from two or more of the ATC categories (except 54% of the African countries). Figure 1A shows main absence of subcategories in different geographic regions. The Western Pacific, eastern Mediterranean and African regions most frequently excluded various subcategories. The two latter regions had the highest number of countries without any endocrine therapy (46% and 30%, respectively). While 12 (80%) of the countries in the region of the Americas had at least one taxane on the list, over 80% in the African (n = 21) and over 90% in the Western Pacific (n = 11) regions had not included any. Monoclonal antibodies were also not included in >50% of the studied European and American countries, as well as over 75% of the countries in all other regions. Similarly, >75% of the countries in all regions (except Europe) did not have any protein kinase inhibitors listed as essential medicine. The median (IQR) number of essential oncology medicines differed between low- (n = 11 [13]), lower-middle (n = 18 [19]) and upper-middle income (n = 26 [34]) countries, although statistical significance was just not reached (Table 1). Trends were similar for antineoplastic agents and endocrine therapy. At least 75% of the low-income and half of the middle-income countries did not designate any taxanes as essential oncology medicines. No difference was observed across income levels when inclusion of monoclonal antibodies, protein kinase inhibitors and ‘α-interferons’ was considered. Notable absent subgroups in different income levels are given in Figure 1B. Endocrine therapy was not selected by 35% of low-income countries. For ‘hormones and related agents’, observed absence was 85% in low-income countries versus 61% in middle-income countries. Countries with >12% cancer mortality had more listed oncology essential medicines than countries with ‘20% 100
“Plant alkaloids and other natural products”
Percentage of absence
90 80 70
“Cytotoxic antibiotics and related compounds”
60 50
“Other antine oplastic agents”
40
“Hormones and related agents”
30
“Hormone antagonists and related agents”
20 10 0 Low income
Lower middle incomeUpper middle income
Figure 1. (A) Absence of subcategories of oncology medicines on NEMLs in WHO regions. (B) Absence of subcategories of oncology medicines on NEMLs in income levels.
Table 2. Oncology essential medicines for different disease burden groups of countries
Cancer mortality as percentage of total mortality Cancer morbidity as percentage of total morbidity
Country groups
Number of countries
Mean oncology essential medicines on the list
SD
P value (ANOVA)
P value (ANOVA)
12% 10%
25 30 21 28 25 23
12 16.5 27 12.3 15.7 27.2
11.8 12.6 14.8 9.5 12.9 16
reference 0.447 0.001 reference 0.623 0.001
– reference 0.022 – reference 0.028
were similar for cancer morbidity measures as well as subcategories (antineoplastic agents and endocrine therapy) of medicines (Table 2). Figure 2A and B shows the association between the number of essential oncology medicines and mortality and morbidity fractions due to cancer. Regression coefficients were 0.78 and 2.52, respectively (P < 0.001 and P = 0.007, respectively), which means an increase in cancer mortality or morbidity fraction by
Volume 25 | No. 1 | January 2014
10% is associated on average with an addition of 8 or 25 oncology medicines on the corresponding NEML, respectively. A total of 52%, 80% and 95% of the countries with 12% mortality due to cancer have chosen their medicines from two or more ATC categories. A similar trend was observed in the clusters of countries grouped by cancer morbidity, where the respective figures were 54%, 84% and 91% of the countries.
doi:10.1093/annonc/mdt514 |
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
Af
er
ic
ric
a
n
“Other antineoplastic agents” a
Percentage of absence
A
original articles
Annals of Oncology
number of oncology essential medicines
A 60 50 40 30 20 10
2%
4%
6%
8%
10%
12%
14%
morbidity (DALYs) fraction (cancer morbidity/all causes morbidity)
number of oncology essential medicines
B 60 50 40 30 20 10 0 0%
5%
10%
15%
20%
25%
30%
35%
40%
moratlity fraction (cancer mortality/all causes mortality) Figure 2. (A) Cancer morbidity (DALYs) fraction versus number of essential medicines per country. (B) Cancer mortality fraction versus number of essential medicines per country.
discussion Overall, the median number of selected essential oncology medicines was 16, with NEMLs being dominated by antineoplastic agents. Five countries did not select any oncology medicine. The overall number and diversity of essential oncology medicines increased by improvement of the income level of the countries and increase in cancer burden and varied across regions. Surprisingly, subcategories of antineoplastic agents frequently used as routine chemotherapy agents were not found among essential medicines in a notable fraction of the studied countries. This category includes medicines which have extensive indications in various types of malignancies. The presented results should not be interpreted solitarily. Cancer therapy might be provided in different settings in LMICs and might not be restricted to (essential) medicines. However, the sustainability of such provisions needs to be scrutinized. The majority of the LMICs did not have any monoclonal antibodies, protein kinase inhibitors or ‘α-interferon’ on their NEMLs, neither had over half of them any taxanes which may imply a lag in adopting newer therapies. This might have changed recently by introduction of generic taxanes and may be improved further by invasion of biosimilars in the near future. Carefully conducted comparative studies are required for adopting such therapies in LMICs considering all controversies including the added clinical value and their cost-effectiveness [20].
| Bazargani et al.
Besides, acquiring new therapies without paying adequate attention to its requirements, might not utilize resources in the most efficient manner [21]. Personalized medicine offers considerable benefits to patients and health care systems by distinguishing between individual responses to a treatment in terms of effectiveness and safety. HER-2 expression and KRAS mutation tests which have high predictive values for trastuzumab and cetuximab therapies are two examples among others [22]. It is worth investigating if LMICs which encompassed new therapies have considered prerequisites of such therapies in their guidelines and their routine clinical practice. Absence of endocrine therapy was observed more in lowincome countries which seems worrisome as these medicines are extensively used in certain types of breast and prostate cancer [18, 19]. Even in the region of the Americas where according to our study hormone antagonists are well presented in NEMLs, empirical evidence has indicated lack of access to aromatase inhibitors—the first choice of endocrine therapy in hormone receptor positive breast cancer particularly in postmenopausal women—in the public sector of upper middle-income countries [23, 24]. Prostate cancer, however, is most abundant in developed countries [25]. LMICs have probably prioritized among different cancer types which might explain the absence of gonadotropin-releasing hormone agonists on many NEMLs. When countries were grouped according to cancer burden a trend was observed in which countries suffering more from
Volume 25 | No. 1 | January 2014
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
0 0%
original articles
Annals of Oncology
Volume 25 | No. 1 | January 2014
treatment of breast cancer. This is believed to contribute to the observed trend in reduction in the mortality rates [23]. On this note, we conclude that allocation of resources should be equally directed towards medicines as well as other required medical technologies and interventions, infrastructures and skilled medical staff (e.g. trained oncologists). In conclusion, this study provides evidence that the selection of oncology medicines on NEMLs is suboptimal in LMICs which consequently may impact access to cancer treatments. This finding does not only apply to newer medicines, it also includes routine chemotherapies and in particular hormone therapies. Countries in which cancer is more burdensome have included more oncology medicines and diversified their NEMLs further compared with other countries. Adequate attention from the national health care decisionmakers and international organizations is deemed necessary to provide a sufficient amount of oncology medicines to LMICs in hope to reduce death toll due to cancer in an equitable manner.
disclosure The authors have declared no conflicts of interest.
references 1. Bloom DE, Cafiero ET, Jané-Llopis E et al. The Global Economic Burden of Noncommunicable Diseases. Geneva: World Economic Forum 2011 (www. weforum.org/EconomicsOfNCD). 2. National Cancer Institute. NCI funded research portfolio; http://fundedresearch. cancer.gov/search/funded?Action=full&fy=PUB2011&type=sic (18 November 2012, date last accessed). 3. World Health Organization. Cancer fact sheet. 2013; http://www.who.int/ mediacentre/factsheets/fs297/en/ (10 August 2013, date last accessed). 4. Clegg LX, Reichman ME, Miller BA et al. Impact of socioeconomic status on cancer incidence and stage at diagnosis: selected findings from the surveillance, epidemiology, and end results: National Longitudinal Mortality Study. Cancer Causes Control 2009; 20(4): 417–435. 5. Denny L. Cervical cancer treatment in Africa. Curr Opin Oncol 2011; 23(5): 469–474. 6. Zucca E, Rohatiner A, Magrath I et al. Epidemiology and management of lymphoma in low-income countries. Hematol Oncol 2011; 29(1): 1–4. 7. World Health Organization. National cancer control programmes. 2013; http ://www.who.int/cancer/nccp/en/ (25 July 2013, date last accessed). 8. Bygbjerg IC. Double burden of noncommunicable and infectious diseases in developing countries. Science 2012; 337(6101): 1499–1501. 9. van den Ham R, Bero L, Laing R. The World Medicines Situation 2011; selection of essential medicines; http://apps.who.int/medicinedocs/en/m/abstract/ Js18770en/ (4 December 2012, date last accessed). 10. World health Organization. Country profiles and monitoring of the pharmaceutical situation in countries; http://www.who.int/medicines/areas/coordination/ coordination_assessment/en/index.html (23 November 2012, date last accessed). 11. World Health Organization. Comparative table of medicines on the WHO Essential Medicines Listfrom 1977–2011; http://www.who.int/selection_medicines/list/en/ (4 December 2012, date last accessed). 12. World health Organization. Essential Medicines Lists and Formularies; http://www. who.int/selection_medicines/country_lists/en/index.html (23 November 2012, date last accessed). 13. Wang L, Ma E, Xu W. Comparative Analysis of China National & Twenty-two Selected Provincial Essential Medicine Lists to the WHO 2011 Model List; http ://apps.who.int/medicinedocs/en/m/abstract/Js18851en/ (23 November 2012, date last accessed). 14. WHO collaborating center for Drug Statistics Methodology. ATC/DDD Index 2012; http://www.whocc.no/atc_ddd_index/ (23 November 2012, date last accessed).
doi:10.1093/annonc/mdt514 |
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
cancer burden selected more essential oncology medicines and diversified their choices from different categories. This might indicate that decision making procedures are reasonably informed about the national health care priorities. However, our regression model suggests that variation in selecting oncology essential medicines can only to a limited extent be explained by variation in cancer burden (very low R 2), thus further studies are required to assess which selection decisions are influenced by disease burden on a national scale. One might even suggest that an overall package of anticancer medicines of a country can be reasonably well balanced despite all the exclusions, taking national cancer incidence patterns into account [26]. There were some limitations in our study. First, medicines complementary to chemotherapy or used for supportive and palliative care were beyond the scope of our study. Owing to the late presentation of cancer in LMICs, palliative care medicines get further priority and should be carefully positioned in a thorough overview. Secondly, we studied NEMLs at a national level, but oncology medicines might also be selected at a subnational level, e.g. in hospital formularies or therapeutic guidelines. Another limitation was the inability to measure actual availability of oncology medicines. Direct measurement of availability and affordability of oncology medicines has not been studied so far in LMICs. Perhaps lack of information and transparency particularly in the oncology field where both inpatient and outpatient data are required explains the absence of such data. Moreover, due to constant lack of health care resources in the public sector, one might argue that a situation analysis in the private sector is also crucial to provide a comprehensive insight. Substantial disparities in access and quality of treatments have been reported in both targeted therapies as well as chemotherapies among the two sectors [24]. Besides, due to massive discrepancies in socioeconomic status of different walks of life in many LMICs, access to medicines and care is not consistent within a country, which makes it difficult to draw an integrated conclusion even at a national level [27]. It is arguable that cancer morbidity and mortality data used in this analysis might not reflect the actual situation due to lack of information, lack of patient registries and underdiagnoses in LMICs. Nevertheless, the WHO data on burden of disease are the most extensive available data which have been reliably used in similar global and regional studies [28]. Lastly, medicines are just one component of care. Treatment of cancer is a multifactorial issue and requires an extensive network of screening, diagnosis, monitoring and different treatment strategies (e.g. surgery, radiotherapy) alongside medicines. There is no evidence to suggest that these components—which are even more resource intensive—are convincingly available in LMICs in an equitable manner [5, 6, 29]. One might argue that constrained health care resources should be allocated with a comprehensive approach to fulfill all the components of cancer treatment. Despite such a seemingly reasonable universal recipe for all types of cancers, there are some contrary evidence-based opinions in case of breast cancer suggesting superiority of treatment over screening in resource limited countries [30, 31]. Similar arguments might be applicable to different situations and cancer types and hence worthwhile investigating further. Mexico can be regarded as a learning case, where health policy reform has led to improved access to screening, diagnosis and
original articles
23. Chavarri-Guerra Y, Villarreal-Garza C, Liedke PE et al. Breast cancer in Mexico: a growing challenge to health and the health system. Lancet Oncol 2012; 13(8): e335–e343. 24. Lee BL, Liedke PE, Barrios CH et al. Breast cancer in Brazil: present status and future goals. Lancet Oncol 2012; 13(3): e95–e102. 25. American Cancer Society. Global Cancer Facts & Figures. 2011; http://www. cancer.org/acs/groups/content/@epidemiologysurveilance/documents/document/ acspc-027766.pdf (20 December 2012, date last accessed). 26. Sikora K, Advani S, Koroltchouk V et al. Essential drugs for cancer therapy: a World Health Organization consultation. Ann Oncol 1999; 10(4): 385–390. 27. Pedraza AM, Pollan M, Pastor-Barriuso R et al. Disparities in breast cancer mortality trends in a middle income country. Breast Cancer Res Treat 2012; 134 (3): 1199–1207. 28. Patel AR, Prasad SM, Shih YC et al. The association of the human development index with global kidney cancer incidence and mortality. J Urol 2012; 187(6): 1978–1983. 29. Agarwal G, Ramakant P, Forgach ER et al. Breast cancer care in developing countries. World J Surg 2009; 33(10): 2069–2076. 30. Autier P, Boniol M, Gavin A et al. Breast cancer mortality in neighbouring European countries with different levels of screening but similar access to treatment: trend analysis of WHO mortality database. BMJ 2011; 343: d4411. 31. Gøtzsche PC, Nielsen M. Screening for breast cancer with mammography. Cochrane Database Syst Rev 2011; (1): CD001877. DOI:10.1002/14651858. CD001877.pub4.
Annals of Oncology 25: 276–282, 2014 doi:10.1093/annonc/mdt524
The prognostic significance of left ventricular ejection fraction in patients with advanced cancer treated in phase I clinical trials R. Said1, J. Banchs2, J. Wheler1, K. R. Hess3, G. Falchook1, S. Fu1, A. Naing1, D. Hong1, S. Piha-Paul1, Y. Ye1, E. Yeh2, R. A. Wolff1 & A. M. Tsimberidou1* Departments of 1Investigational Cancer Therapeutics; 2Cardiology; 3Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, USA
Received 16 January 2013; revised 5 September 2013; accepted 16 October 2013
Background: New targeted agents may cause acute cardiac events. The purpose of our study was to investigate the incidence and the prognostic significance of left ventricular ejection fraction (LVEF) in phase I trials. Patients and methods: Between October 2008 and September 2011, the records of 1166 consecutive patients with advanced cancer treated in the Phase I Clinic who underwent echocardiography were retrospectively reviewed. Results: Most of the patients were White (78%), and the most common tumor types were colorectal cancer and melanoma. Of 1166 patients, 177 (15.2%) patients had an LVEF of 35%, ≤2 prior systemic therapies, ≤2 metastatic sites, and normal lactate dehydrogenase and albumin levels. Conclusion: Echocardiography would improve patient selection for enrollment in phase I clinical trials. These data suggest that it is safe to treat patients with LVEF between 35% and 50%. Key words: biomarker, cardiac dysfunction, malignancy and phase I trial *Correspondence to: Dr A. M. Tsimberidou, Department of Investigational Cancer Therapeutics, The University of Texas MD Anderson Cancer Center, Unit 455, 1515 Holcombe Boulevard, Houston, TX 77030, USA. Tel: +1-713-792-4259; Fax: +1-713794-3249; E-mail: [email protected]
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at University Library, University of Illinois at Chicago on June 22, 2014
15. World health Organization. Health Statistics and Health Information Systems; Disease and injury country estimates; http://www.who.int/healthinfo/global_ burden_disease/estimates_country/en/index.html (23 November 2012, date last accessed). 16. World health Organization. WHO geographic regions; http://www.who.int/about/ regions/en/index.html (23 November 2012, date last accessed). 17. The World Bank. World Bank classification of countries; http://data.worldbank.org/ indicator/NY.GNP.PCAP.CD (23 November 2012, date last accessed). 18. National Cancer Institute. Breast cancer treatment; treatment options overview; http://www.cancer.gov/cancertopics/pdq/treatment/breast/Patient/page5#Keypoint 24 (5 December 2012, date last accessed). 19. National Cancer Institute. Prostate cancer treatment; treatment options overview; http://www.cancer.gov/cancertopics/pdq/treatment/prostate/Patient/ page4 (5 December 2012, date last accessed). 20. Greenberg D, Earle C, Fang CH et al. When is cancer care cost-effective? A systematic overview of cost-utility analyses in oncology. J Natl Cancer Inst 2010; 102(2): 82–88. 21. Stolk P, Willemen MJ, Leufkens HG. Rare essentials: drugs for rare diseases as essential medicines. Bull World Health Organ 2006; 84(9): 745–751. 22. Blair E. Predictive tests and personalised medicine. Drug Discov World 2009; (Fall): 27–31. http://integratedmedicines.co.uk/images/presentations/E_Blair_ Predictive_tests.pdf.
Annals of Oncology
