Inequalities in lung cancer: a world of EGFR

Discussion

The results of the present study covered 78% of the worldwide population. We showed that EGFR mutational analysis was available for 70% of the worldwide population and for 42.6% thereof at a mean cost of <US$500. Erlotinib was the most widely available EGFR-TKI, accessible for 75% of the worldwide population and for 30% thereof at a mean cost of <US$500. We found a clear association between a low/medium HDI rank and the unavailability of EGFR testing or EGFR-TKIs. Similarly, we illustrated an association between low/medium HDI rank and higher costs for tests or treatments.

To the best of our knowledge, our study is the first of its kind. Yatabe et al. [11] published a retrospective study involving 40 sites in 11 Asian-Pacific countries. Their objective was to determine the proportion of patients who benefited from EGFR analysis in routine practice in 2011. They reviewed 22 000 medical files and reported that 31.8% of the patients had been tested for EGFR mutations. Although they found disparities between countries (64.8% of patients tested in Japan versus 18.3% in China), they performed no comparisons. Other retrospective studies have focused on EGFR mutation risk factors, patient eligibility for the analysis, mutation analysis techniques and the application of national therapeutic recommendations when they exist [12–14], but these efforts were uniquely national in scope, with no international comparisons.

Our present study illustrated differences in patient access to diagnostic tests and their associated treatments as a function of country development level. These differences may have several explanations, including not only the specificity of the equipment and knowledge needed for mutation testing, but also delays in the transfer of competencies to lesser developed countries. Furthermore, the costs of the techniques and the deployment of networks require substantial investments at the national level. Another limiting factor is cost remaining for the patient. Indeed, our study illustrated that only very highly developed countries provide these analyses and treatments with no direct out-of-pocket expenses for the patient. This problem surely limits the development of and access to innovative practices and treatments, even though early diagnosis and surgery remains the best way to cure lung cancer [15]. Finally, despite the fact that lung cancer causes more deaths than any other cancer worldwide and that 80% of these deaths occur in low-resource countries, the malignancy is not a public health priority in numerous underdeveloped countries, where infectious diseases remain the leading cause of mortality [16, 17].

As is the case for all specialist opinion surveys, our study does have limitations. First, we received responses from only 74 of the 194 UN Member States, which creates a selection bias in our study. We did, however, demonstrate that most of the nonresponding countries had a low HDI. Furthermore, many had a political context poorly suited to responding to this type of survey (see online supplementary material) and many had small populations. Given the low HDI rank of the nonresponding countries, we feel that it is unlikely that they have this analysis routinely available and consequently it is unlikely that our results are underestimated. Second, our survey is based on responses provided by specialists, which introduces an interpretation bias in our study. Indeed, it is possible that some of the responding specialists were unfamiliar with the situation at the national scale, noting that for some countries we had difficulties just finding someone willing to respond to our survey. We were pleasantly surprised, however, to observe that in countries where there were multiple responding specialists, these latter usually provided comparable responses concerning the most important issues, e.g. the availability of EGFR testing and EGFR-TKIs. Moreover, costs should be interpreted with caution since experts may have misinterpreted the order to retain remaining cost for the patient and not the global cost (i.e. for patient and public health insurance/system). However, our results are linked to the HDI, which may be interpreted as an indirect evidence of data quality. Third, there is likely variability in the availability (and the cost) of EGFR testing and EGFR-TKIs in nations functioning on a federal model, within which subnational divisions have autonomy in health provision. This is all the more so a problem in that it often concerns nations with large populations (USA, India, Brazil, China, etc.). Results may also be biased by the type of health insurance system in each country (ranging from fully government to fully private) which is not assessed by the HDI. Moreover, some other factors, not recorded in this study, may influence EGFR-TKI availability and/or cost, such as reimbursement status or local formulary approval. For EGFR testing, the method of testing (panel versus single gene) may also impact our results. Fourth, our study is a cross-over survey (conducted between April 2014 and November 2014). Thus, some data may have been outdated when published. One strong example is the recent approval of gefitinib for the USA [18].

Despite these limits, our study does provide a glimpse of the current, worldwide availability of EGFR mutation testing and associated treatments. We hope that our results can be a launching point for the creation of a worldwide network to implement and follow-up upon such mapping in the future.

Our study also has strengths. To the best of our knowledge, it is the first to assess the availability of EGFR mutation testing and resulting treatments at a worldwide level. Furthermore, we used a very simple questionnaire that could be sent by E-mail and completed rapidly. It comprised single-response questions, from which homogenous and comparable data were collected. With this questionnaire, we were able to collect responses from 100 specialists capable of describing the situation for 78% of the worldwide population. Our study also benefited from the support of several international academic associations (ERS, APSR, ALAT, and the Thoracic Society of Australia and New Zealand), which strengthened its reach and stature. To the best of our knowledge, the present study is also the first to consider its end-points in comparison with a reliable index of national development.

Thus, 10 years after the arrival of molecular analysis and targeted therapies in thoracic oncology, our study is the first to underline the lack of access to these undeniably efficacious technologies and treatments [3] in low HDI countries.

The data retrieved from our survey enabled an unprecedented cartography of worldwide access to EGFR molecular analysis techniques and EGFR-TKIs. Therefore, in 2014, essentially in poor countries, a significant number of patients may have been candidates for these highly effective treatments, but incapable of receiving them. Participating in clinical trials, developing generics, deploying molecular genetics networks, creating public–private partnerships, and strengthening the roles of international institutions and academic associations are all possibilities that must be explored to improve access to these effective but costly innovative technologies.