Is It Time To Ask a Disturbing Question About The Geographical Location of Cancer Clinical Trials?

Posted by Samantha Powell on June 25th, 2014 |

The genesis of this commentary is the recent publication of population-based survival outcomes associated with specific malignancies in various European countries.1 This analysis, which covered the years 1999 to 2007 revealed rather striking variation in the survival of patients with the specific cancers based on geographical location.

For example, five-year survival in the United Kingdom (England) for stomach cancer was only 17%, whereas in Iceland, Austria, Belgium, Germany, Italy, Switzerland and Portugal, survival for this patient population at the same time point was greater than 30%. Similarly, for kidney cancer, five-year survival in the United Kingdom was only 47.3%, whereas the same figure was at least 10% higher in each of the previously noted countries as well as France, Spain, Finland and several Eastern European countries.

Of course, critical questions to be addressed are why do these clearly nontrivial differences in cancer survival exist between countries with apparently similar ethnic, socioeconomic and even political backgrounds, and what can be done as quickly as possible to improve what can only be appropriately described as “concerning” survival disparities?

The specific intent of this commentary is not to highlight this general issue, as important as it is to the health and welfare of the citizens of certain countries, but rather to inquire about the potential major impact of these observations on the interpretation of cancer clinical trial results that might be reported in particular cancer types that originate from geographical areas noted to have particularly inferior survival.

For example, the same issue of the major oncology journal reporting the European cancer survival data reported the results of a Phase III randomized trial that had examined the clinical utility of docetaxel plus “active symptom control” compared with “active symptom control” alone for “refractory esophagogastric carcinoma.”2This study, conducted in the United Kingdom, found a statistically significant improvement in overall survival (OS) (median 5.2 vs. 3.6 months; hazard ratio, 0.67;P=0.01) associated with inclusion of docetaxel antineoplastic therapy.

The results of this trial clearly are welcome news for patients diagnosed with gastric cancer who may require treatment in this setting. However, one must ask if the same study outcome would have been observed if the trial had been conducted in a country where routine “standard of care” (perhaps including the delivery of docetaxel or alternative potentially beneficial second-line antineoplastic treatments) might have led to a superior OS outcome in the “control population.”

Although clearly a complex issue and one requiring far more discussion than permitted in this brief commentary, the striking differences in cancer survival observed in the previously noted population-based report may directly relate to the availability of specific antineoplastic strategies in certain countries and not others. And if a given approach is available as a routine component of standard of care (including particularly following a patient’s removal from a clinical trial due to disease progression) in one country but not in another, in the opinion of this commentator, it is highly possible (or even probable) that variations in OS between study arms could occur solely due to this non-study–related factor (based on the geographical location), similar to what is observed outside the study setting.

Finally, how does this factor influence the critical issue of the generalizability of a given study outcome when it was conducted in a setting where the population of patients with a specific cancer type is recognized to experience inferior OS? Although clearly a tough question, the issue surely needs to be considered.

Date: 6/25/14
Outlet Full Name: Clinical Oncology
Author: N/A

Comments are closed.