Annals of Oncology Advance Access published online on October 8, 2009
Annals of Oncology, doi:10.1093/annonc/mdp403
Global drug development in cancer: a cross-sectional study of clinical trial registries
Division of Medical Oncology and Hematology, Princess Margaret Hospital and University of Toronto, Toronto, Ontario, Canada
* Correspondence to: Dr I. F. Tannock, Division of Medical Oncology and Hematology, Princess Margaret Hospital, 610 University Avenue, Toronto, Ontario, Canada M5G 2M9. Tel: +1-416-946-2245; Fax: +1-416-946-4563; E-mail: ian.tannock{at}uhn.on.ca
 |
abstract |
|---|
 Top abstract introductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
Background: Drug development in cancer is costly and may be directed toward ‘profitable’ cancers of the more developed regions (MDR) as compared with those of the less developed regions (LDR) of the world. Here, we describe drug development in relation to cancer type and geographic location.
Materials and methods: We reviewed phase II and III clinical trials evaluating new cancer drugs, which were registered from January to June 2008. Correlations were sought between the number of clinical trials and incidence, mortality and prevalence of the cancers studied (obtained from GLOBOCAN 2002) and stratified by region of the world.
Results: We identified 399 newly registered trials. Most trials (N = 229, 57%) were sponsored by industry. The most common types of cancer studied were breast 73 (18%), lung 57 (14%), prostate 44 (11%) and colorectal 28 (7%). In MDR, incidence, mortality and prevalence correlated significantly (Pearson r = 0.80, 0.73 and 0.63; P 
0.01) with the number of all registered clinical trials, whereas in LDR, only prevalence showed significant association (Pearson r = 0.55; P = 0.03) with the number of trials for a given type of cancer.
Conclusion: Lethal cancers that are common in the LDR (e.g. stomach, liver and esophageal cancers) deserve greater emphasis for drug development.
cancer, drug development, global
|
introduction |
|---|
|
Topabstractintroductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
The world population in the year 2000 exceeded 6 billion with only 20% of the population living in the more developed regions (MDR) of the world (i.e. North America, Europe, Australia, New Zealand and Japan as defined in GLOBOCAN 2002) [1] and the rest living in the less developed regions (LDR) [2]. The global burden of cancer is increasing concurrently with the expanding world population. According to the GLOBOCAN database, an estimated 11 million new cancer cases (i.e. incidence) and 6.7 million cancer deaths (i.e. mortality) occurred worldwide in 2002. The highest incidence sites of cancer worldwide were lung, breast, colorectum, stomach and prostate, whereas the highest mortality was associated with lung, stomach, liver, colorectal and breast cancers; breast, colorectal and prostate cancers were the most prevalent cancers globally (Table 1) [1, 3]. Estimated global burden of cancer in 2004 expressed as disability-adjusted life-years (DALYs) lost closely follows the pattern of mortality [4], and this loss is projected to increase in the LDR but not the MDR (Table 2).
|
|
Improved knowledge of the biology of cancer has led to the development of new anticancer drugs, which are evaluated in clinical trials. Phase II and III clinical trials study efficacy and clinical benefit (as well as toxicity), respectively, and are almost always cancer specific, i.e. a particular trial evaluates only one type of cancer. To avoid publication bias and duplication of trials, and to better plan future research, the International Committee of Medical Journal Editors (ICMJE) has established registration of all new clinical trials with a penalty of denial of publication in participating journals should trials fail to be registered [5, 6].
Sponsorship of phase III clinical trials in oncology has shifted substantially over the last three decades, from government sources to the pharmaceutical industry [7]. The cost of bringing a new cancer drug to the market is estimated at 
$1 billion [8]. Market prospects and return on costly investments dictate drug development in common nonmalignant illnesses of wealthier countries and the relatively minimal drug development for common infectious diseases of poorer countries [9]. In concurrence with this, 80% of the world’s total financial resources devoted to health care are spent currently in wealthy countries [10]. In oncology, high costs of drug development lead inevitably to attempts to maximize profit in the marketplace.
We hypothesize that clinical development of new cancer drugs predominates for potentially more profitable cancers, which are the most lethal and prevalent cancers in wealthier countries in contrast to ‘less profitable’ cancers which cause high mortality and loss of DALYs in poorer countries. In the present cross-sectional study, we evaluated phase II and phase III clinical trials registered in Primary and Partner Registries of the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) to evaluate drug development in relation to cancer type and geographic location.
|
materials and methods |
|---|
|
Topabstractintroductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
search strategy and study selection
We conducted a search in six Primary Trial Registries: Australian New Zealand Clinical Trials Registry, Australia/New Zealand; ISRCTN.org, UK; Netherlands National Trial Register, The Netherlands; China Clinical Trial Register, China; Clinical Trials Registry-India, India and Sri Lanka Clinical Trial Registry, Sri Lanka and two Partner Registries: Physician Data Query (affiliated registry ClinicalTrials.gov, United States) and European Leukemia Trial Registry (affiliated registry German Clinical Trial Register, Germany), which are accessible electronically through the WHO ICTRP [11]. WHO Primary and Partner Registries must meet criteria in the following categories to assure that the data collected meet the standards of the ICTRP: content, quality and validity, accessibility, unique identification, technical capacity and administration and governance. A minimum amount of information (20 items) is required for a trial to be fully registered (Table 3). The registries were chosen because they meet the standard eligibility criteria of the ICMJE and the WHO and for the convenience of being available through a single search portal. ICTRP was accessed in July 2008. Using the advanced search portal, we entered the terms ‘cancer’ or ‘neoplasm’ in the ‘condition’ box, leaving the ‘title and intervention’ boxes empty. The date range was set from 1 January to 30 June 2008. All other fields were left blank or selected as ‘ALL’; the latter refers to inclusion of all Primary and Partner Registries present in the ICTRP at the time of the search.
|
data extraction
A data extraction form was designed to capture information related to the studies. Data were collected regarding characteristics of the study (site of recruitment, registration information, phase of clinical trial and setting), features of the intervention (chemical and trade names of the agent) and the category of the agent. Sponsorship was determined using established criteria [12–14]. Any information not presented directly in the search portal was sought using a Google search. The extraction form was piloted on 20 trials and modified as necessary. All data were abstracted and entered into a database using Microsoft Access by a single author (PCH).
geographic categories and cancer incidence, mortality and prevalence
To identify regional disparities in recruitment into phase II and III clinical trials, we used two geographic categories defined in GLOBOCAN 2002 [1]: (i) MDR (i.e. all regions of North America, Europe, Australia, New Zealand and Japan) and (ii) LDR (i.e. the rest of the world). Of note, some developed parts of Asia (e.g. Hong Kong and Singapore) are not included in the MDR. Clinical studies recruiting in both MDR and LDR were categorized as worldwide. Estimates of cancer incidence, mortality and prevalence in 2002 were obtained from GLOBOCAN 2002 [1] and from peer-reviewed publications, which are based on the GLOBOCAN and Cancer Incidence in Five Continents databases [3, 15].
statistics
Descriptive statistics was used to summarize the studies by region and by cancer type. We selected 15 types of cancer with the largest number of registered clinical trials overall in the analysis. Logarithmic transformation of variables was carried out to achieve approximate normal distribution of variables before analysis. Pearson correlation coefficients were used to evaluate the relationship between the number of registered clinical trials and incidence, mortality and prevalence in the MDR and LDR: our primary hypothesis addressed the relationship with mortality. All analyses were carried out using SAS v9.1 (SAS Institute, Cary, NC). Statistical significance was considered at P <0.05 for evaluation of the primary hypothesis; P values for relationships with incidence and prevalence are derived without correction for multiple comparisons.
|
results |
|---|
|
Topabstractintroductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
search results
Our search yielded a total of 1594 trials. This list of trials was manually filtered and ineligible trials were excluded according to the following criteria: phase I/pilot or phase IV/long-term extension studies (n = 451), studies evaluating surgery or radiation modalities (n = 202), trials dealing with screening and prevention (n = 417) and trials evaluating nonmedical therapies (e.g. psychosocial support) (n = 125).
general characteristics of clinical trials
From January to June 2008, 399 phase II and phase III clinical trials were newly registered in registries based in the MDR: 375 (94%) trials in ClinicalTrials.gov, 18 (5%) trials in ISRCTN and 6 (2%) trials in ANZCTR; no trials were double registered. There were no eligible trials registered in the three registries based in the LDR. Of 374 clinical trials with available information about recruitment, 322 (86%) recruited participants only in the MDR, 39 (10%) only in the LDR, and 13 (3%) worldwide. Altogether, 324 (81%) clinical trials were phase II and only 75 (19%) were phase III. In 229 clinical trials (57%), the pharmaceutical industry was listed as a source of sponsorship (i.e. for profit and mixed sponsorship), whereas other trials were funded from nonprofit sources (i.e. governments, foundations, cooperative groups and hospitals).
clinical trials and types of cancer
The most common types of cancer studied in phase II and III clinical trials in our cohort were breast 73 (18%), lung 57 (14%), prostate 44 (11%) and colorectal 28 (7%). Similarly, the highest number of phase III clinical trials evaluated drugs for breast (15), prostate (15), lung (11) and colorectal (8) cancers (Table 4).
|
There was a strong association between incidence, mortality and prevalence in the MDR and the number of all registered clinical trials for the 15 sites of cancer with the highest number of registered clinical trials (Pearson r = 0.80, 0.73 and 0.63; P < 0.001, 0.002 and 0.01, respectively). In contrast, in the LDR, only prevalence showed a significant association with the number of trials for a given type of cancer (Pearson r = 0.55; P = 0.03), whereas associations with incidence and mortality were weak (Pearson r = 0.48 and r = 0.38; P = 0.07 and P = 0.17, respectively). Secondary analyses of the associations between incidence, mortality and prevalence in the MDR and LDR and the number of clinical trials recruiting in MDR and LDR only showed similar associations; however, in the LDR, clinical trials were recruiting in only eight cancer sites (Table 4) and therefore, the results of this secondary analysis should be interpreted cautiously.
clinical trials and types of treatment
Targeted therapies with small molecules or monoclonal antibodies (mAbs) were studied in 220 (55%), chemotherapy in 97 (24%), immunotherapy (not including mAbs) in 50 (13%), hormonal therapy in 18 (5%) and other therapies in 14 (4%) clinical trials. Targeted therapies were studied as monotherapy in 81 (37%) of the clinical trials and in combination with chemotherapy and other therapies (e.g. hormonal therapy and other targeted therapies) in 99 (45%) and 40 (18%) of the clinical trials, respectively. The highest number of different categories of targeted therapies was studied in prostate (10), breast (9), lung (7), colorectal (6), ovary (6), stomach (6), liver (5) and pancreatic (5) cancers.
|
discussion |
|---|
|
Topabstractintroductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
We found a significant and strong association between global clinical drug development and cancer mortality in the MDR but no significant association with cancer mortality in the LDR. There were also strong and significant associations between number of trials and incidence and prevalence of cancers in the MDR but only a weak association with prevalence of cancers in the LDR. This is not surprising since the most prevalent cancers in the MDR are also among the most prevalent in the LDR (Table 1), but the cancers responsible for highest levels of mortality and loss of DALYs differ. The observations of greater clinical drug development for lethal (and highly prevalent) cancers of the MDR (i.e. lung, breast, colorectal and prostate cancers) as compared with highly fatal cancers of the LDR (e.g. stomach, liver and esophageal cancers) may reflect profit-driven global drug development in cancer. Drug development for highly incident and prevalent cancers of wealthy countries may assure a higher chance to bring a new drug to market.
Mortality is the most important measure of cancer burden and progress against cancer [16] and it is expected to decrease in the MDR with improvements in prevention, screening or treatment [17]. In 2002, 40% of all cancer deaths occurred in the MDR [15], where only 20% of the global population lives [2]. In our analysis, global drug development was strongly associated with cancer mortality only in the MDR, where the most incident and prevalent cancers are also the major killers (with the exception of prostate cancer) and represent the highest cancer burden (Tables 1 and 2). Prostate cancer is a less important cause of mortality in the MDR and globally; because it is a disease of older men, fewer ‘healthy’ years of life are lost than for some other sites. In contrast, stomach cancer is a more important cause of mortality than prostate cancer worldwide (Tables 1 and 2) but shorter survival leads to a much lower prevalence. Almost three times fewer (16 versus 44) phase II and III clinical trials are undertaken in stomach than in prostate cancer. From the perspective of the wealthier world, global drug development in cancer might seem appropriate, with some exceptions (e.g. stomach cancer). In contrast, drug development for lethal cancers in less wealthy countries (e.g. stomach, liver and esophageal cancers), where 60% of cancer deaths occur, is under-researched.
Cancer statistics and cancer publications, 90% of which are produced in the western world [18], often emphasize the importance of breast and prostate cancers, which have a chronic course in many patients. According to our analysis, the largest number of clinical trials and the largest number of phase III clinical trials (which have the potential to bring a new drug to market and change clinical practice) are being undertaken in breast and prostate cancers. Moreover, the highest number of targeted therapies is also being evaluated in these cancers.
In agreement with the analysis of Booth et al. [7] for contemporary phase III randomized clinical trials, 57% of all phase II and III clinical trials from our cohort were sponsored by companies. Phase III clinical trials, especially those sponsored by industry, have become larger [7] and have the power to detect small, statistically significant but not necessarily clinically meaningful benefits of new therapies. Expensive new drugs with rather minimal clinical benefit have been added to the already broad armamentarium of therapeutics in breast [19] and colorectal cancers [20, 21]. Concordant with this observation, 68.7% of new chemical entities marketed globally from 1975 to 1999 for various illnesses, which were developed mostly for the needs of wealthy countries, were found to present little or no therapeutic gain compared with those already available [9]. We face intensive clinical drug development of new drugs with small and questionable clinical benefits in major cancers of the MDR, whereas drug development for major cancers in poorer countries is an unmet need.
According to the WHO projections of cancer burden for the year 2030, lung, stomach, esophagus and liver cancers will contribute the highest increase in global cancer burden [4] (Table 2); these projections reflect the incidence of these cancers in the fast-growing population of the LDR. An estimated 35% of cancer mortality has been attributed to well-established modifiable risk factors [22] and primary prevention remains the most important and the most feasible method to reduce cancer burden. However, there is also a need for more intensive drug development to improve treatments for these types of cancer.
The strength of our analysis is in the use of a comprehensive database of valid, accurate, unique and recent information about registered clinical trials. Statistics for the LDR sample might yield more significant results with a larger sample of clinical trials. However, we assume that registration of clinical trials is continuously improving so that a ‘snapshot’ of clinical trial registries for a recent time period is likely to reflect clinical drug development more reliably than analysis over a longer time period. Our analysis is therefore cross-sectional and we were not able to evaluate temporal trends. It is possible that unregistered clinical trials are being conducted in the LDR, although the impact of such trials is likely to be minimal if they are denied publication.
In conclusion, with the exception of lung cancer, which is a global problem, drug development for cancer is focused predominantly on prevalent cancers of the MDR, which are the most important cause of cancer death there but not in the LDR. From a global perspective, a higher chance of bringing a new drug to market should not guide drug development in cancer. Stomach, liver and esophageal cancers will remain major killers globally and, in addition to effective prevention, should receive priority for preclinical and clinical drug development.
|
acknowledgements |
|---|
|
Topabstractintroductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
We thank Monika Krzyzanowska for her helpful comments about an earlier draft of this manuscript.
Received for publication July 7, 2009. Accepted for publication July 10, 2009.
|
references |
|---|
|
Topabstractintroductionmaterials and methodsresultsdiscussionacknowledgementsreferences |
|---|
1. International Agency on Research of Cancer. GLOBOCAN 2002. http://www-dep.iarc.fr/ (8 November 2009, date last accessed).
2. Population in 1999 and 2000: All Countries. United Nations. Population Division: Department of Economic and Social Affairs. http://www.un.org/popin/popdiv/pop1999-00.pdf (10 November 2009, date last accessed). Source: World Population Prospects: The 1998 Revision. Vol. I: Comprehensive Tables (United Nations publication, Sales No. E.99.XIII.9).
3. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin (2005) 55(2):74–108.
4. World Health Organization. Global burden of disease. http://www.who.int/healthinfo/global_burden_disease/en/ (10 January 2009, date last accessed).
5. De Angelis C, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. N Engl J Med (2004) 351(12):1250–1251.
6. De Angelis CD, Drazen JM, Frizelle FA, et al. Is this clinical trial fully registered? A statement from the International Committee of Medical Journal Editors. N Engl J Med (2005) 352(23):2436–2438.
7. Booth CM, Cescon DW, Wang L, et al. Evolution of the randomized controlled trial in oncology over three decades. J Clin Oncol (2008) 26:5458–5464.
8. Adams CP, Brantner VV. Estimating the cost of new drug development: is it really 802 million dollars? Health Aff (Millwood) (2006) 25(2):420–428.
9. Trouiller CP, Olliaro P, Torreele E, et al. Drug development for neglected diseases: a deficient market and a public-health policy failure. Lancet (2002) 359(9324):2188–2194.[CrossRef][Web of Science][Medline]
10. World Health Organization. Spending on health: a global overview. http://www.who.int/mediacentre/factsheets/fs319/en/ (10 January 2009, date last accessed).
11. World Health Organization. International Clinical Trials Registry Platform (ICTRP). http://www.who.int/ictrp/en/ (5 July 2008, date last accessed).
12. Ridker PM, Torres J. Reported outcomes in major cardiovascular clinical trials funded by for-profit and non-for-profit organizations: 2000–2005. JAMA (2006) 295(19):2270–2274.
13. Clifford TJ, Barrowman NJ, Moher D. Funding source, trial outcome and reporting quality: are they related? Results of a pilot study. BMC Health Serv Res (2002) 2(1):18.[CrossRef][Medline]
14. Rochon PA, Gurwitz JH, Simms RW, et al. A study of manufacturer-supported trials of nonsteroidal anti-inflammatory drugs in the treatment of arthritis. Ann Intern Med (1994) 154(2):157–163.[CrossRef]
15. Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol (2006) 24(14):2137–2150.
16. Extramural Committee to Assess Measures of Progress Against Cancer. Measurement of progress against cancer. J Natl Cancer Inst (1990) 82:825–835.
17. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin (2008) 58(2):71–96.
18. Tas F. The contribution of countries and world regions in productivity of oncological publication. Ann Oncol (2008) 19(11):1962–1968.
19. Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med (2007) 357(26):2666–2676.
20. Saltz LB, Clarke S, Diaz–Rubio E, et al. Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal cancer: a randomized phase III study. J Clin Oncol (2008) 26(12):2013–2019.
21. Booth C. Bevacizumab in advanced colorectal cancer: a challenge to the current paradigm. J Clin Oncol (2008) 26(28):4693–4694.
22. Danaei G, Vander Hoorn S, Lopez AD, et al. Causes of cancer in the world: comparative risk assessment of nine behavioural and environmental risk factors. Lancet (2005) 366(9499):1784–1793.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||