In our recent Journal of Clinical Oncology article entitled “Determinants of early mortality among 37,568 patients with colon cancer who participated in 25 clinical trials from the adjuvant colon cancer endpoints database” [6], we reported the results of a study that characterized the determinants of early mortality in a large cohort of early-stage colon cancer patients who had participated in prior adjuvant clinical trials. This study was conducted because the factors associated with early death after surgery and adjuvant chemotherapy are poorly defined. Therefore, we conducted a pooled analysis of over 30,000 patients from 25 randomized clinical trials of adjuvant systemic therapy. Using multivariate logistic regression models and controlling for confounders, we successfully developed and validated a nomogram for 6-month mortality.
We found that early mortality was very low: 0.3% at 30 days, 0.6% at 60 days, 0.8% at 90 days, and 1.4% at 6 months [6]. Consistent with other studies [7–9], our prognostic analyses showed that advanced age, male gender, worse performance status, and higher tumor stage and grade predicted a greater likelihood of early mortality, whereas treatment received was not significantly associated with early mortality. Our findings underscored the observation that early mortality was generally uncommon, but it was more frequently seen in specific subsets of patients, such as those who were older and frailer. This highlights the importance of tools that can better clarify the benefit-to-risk ratio to patients who are considering clinical trial participation. The nomogram developed from our analysis has been validated and calibrated to serve as a potentially effective instrument that can guide and enhance treatment decision-making and discussions between clinicians and patients.
This study is also a proof-of-principle for other countries in terms of illustrating the strength of large databases. Colon cancer is a common cancer worldwide [10]. Particularly in China, the incidence of colon cancer is anticipated to increase significantly over time, and the burden of this disease and its effect on society are expected to grow exponentially, especially given the longer lifespan of patients that has resulted from recent diagnostic and therapeutic advances [11, 12]. Although clinical trials are offered globally, the number of phase III trials available in China often pales in comparison to the number available in Western countries. Reasons for this disparity are numerous and may include various clinical and systemic factors, such as infrastructure and resource constraints or concerns regarding the risks of adverse events that may be more prevalent or unique among Asian patients.
As in many other countries, in China participation in clinical trials is suboptimal. It is estimated that only 2%–4% of patients in China with cancer ultimately consent to enroll in clinical trials, even when studies are available and offered to eligible patients [13]. This finding has been attributed in part to inherent cultural beliefs that regard clinical trials as socially undesirable [14]. Interestingly, these negative perceptions of clinical trials have been shown to dissipate after effective educational interventions and open discussions with physicians [15]. As such, our nomogram may aid clinicians in their conversations with patients.
Because large clinical trials are not always accessible, and since important research questions almost always require adequate sample sizes to address well, pooling data from either population-based settings or clinical trial settings is an effective strategy for cancer outcomes research. Currently, the ACCENT database comprises patient-level data from over 25 adjuvant colon cancer clinical trials from North America and Europe. When the outcome of interest, such as early mortality, is relatively rare, pooling data can add validity to the analysis and strength to its findings. Presently, China is not part of similar database-driven initiatives, but pooling of such data may prove to be an invaluable resource for Asian patients [16]. Databases such as those available at BCCA or via ACCENT are excellent examples of the power of pooled data. The creation of similar databases of Asian patients should be strongly encouraged to make investigations of rare but clinically pertinent endpoints more feasible [17].
