Collaboration between AI-based platforms and health care organizations offers promising results.
In recent years, the landscape of cancer care has undergone remarkable transformations, driven by advances in medical science and technology, particularly through the development of targeted therapies and precision medicine. However, despite the revolutionary advancements in cancer care, ethnic and socioeconomic disparities persist, not only when it comes to accessing and benefiting from these innovations but also in the outcomes experienced by patients.1 For instance, African American patients are known to have higher incidence and mortality rates from certain types of cancer, including colorectal, breast, prostate, and lung; this highlights the unequal burden that certain ethnic groups continue to face in cancer care.2,3
Amid these challenges, the emergence of patient-centric artificial intelligence (AI) decision support systems has begun to offer a promising avenue for addressing existing gaps in health equity in cancer care. Specifically, there are AI platforms that have been developed that use AI to provide a patient-centric treatment decision support opportunity for patients with cancer. Through this technology, patients can input their disease characteristics, treatment history, and general health while also indicating preferences such as location, travel distance, and immediate goals of treatment. Based on the information, the platform generates a personalized and actionable treatment plan that includes both standard and clinical trial options relevant to patients’ preferences and health needs.
Additionally, this patient-oriented approach allows AI-based platforms to identify and resolve barriers encountered by patients with cancer when considering treatment and enables relevant health care and support organizations to effectively address individual patient health care issues. In this way, patient-centric AI platforms can act as supplementary tools to support targeted strategies by health organizations to promote informed patient decision-making and access to cancer trial sites, as the AI platform can also help to guide the strategic establishment of trial sites in areas where potential matched patients are located. This collaboration opportunity between AI platforms and health care organizations offers a seamless approach to mitigate disparities, improve equitable access to clinical trials, and enhance health care outcomes for underrepresented populations, as shown in FIgure 1A.2-5
In a study published in the Journal of Clinical Oncology and presented at the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting, investigators set out to characterize disparities in both disease knowledge and access to cancer care, as well as explore potential mitigation using AI-based digital tools.4 The focus was specifically directed toward genetic testing–related disparities, given its vital importance in identifying targeted therapies for specific genetic makeups and its strong association with improved clinical outcomes, including extended survival.5 Another crucial aspect is the accurate comprehension and interpretation of genetic testing reports, which constitutes an additional critical step toward reaping its benefits.
The study encompassed 11,228 patients in the United States who had advanced cancer. General disease knowledge, as measured by questionnaire completion rate, was high across ethnic groups (P = .16), highlighting the relevance of a patient-driven digital platform for the identification of cancer treatments regardless of ethnic background. Further, the efficacy of the digital platform used in the study to foster engagement within a diverse patient population serves as an encouraging model for inclusive health care access, which is depicted in Figure 1B below.2-5
A noteworthy departure from conventional clinical trial approaches was also observed in the digital platform, as it had an increased representation of non-White patients with cancer. In total, approximately one-quarter of the cohort consisted of non-White patients, which is in sharp contrast to the historical underrepresentation of these demographics in cancer clinical trials.6,7 The platform’s capacity to attract a more diverse patient population highlights its potential to democratize access not only to care but also to inclusion in clinical trials.
The study presented at ASCO extensively investigated the scope of genetic testing across the diverse patient population, revealing distinct patterns of testing disparities based on ethnic backgrounds. African American and Hispanic patients demonstrated significantly lower rates of undergoing genetic testing compared with their counterparts from other groups, which is illustrated in Figure 2A.6,7 Although 47% of the entire cohort underwent genetic testing, the rate was notably lower among African American patients and Hispanic patients at 40%. This pronounced disparity underscores the urgency of addressing multifaceted barriers, including health care access limitations, awareness gaps, and educational disparities that collectively contribute to perpetuating the unequal prevalence of genetic testing.8,9 Further, of the patients who did have genetic testing performed, only 28% provided correct detailed information about the findings in their respective reports, reflecting a clear understanding of their genetic test results. The knowledge gap was most pronounced among African American patients, with only 15% demonstrating and reporting knowledge of their tumors’ mutation status, shown in Figure 2B.6,7
It is also important to recognize that significant knowledge gaps persist in comprehending genetic test results across the entire cohort of patients included in this study. Because of the relative complexity of the report, the significant knowledge gap is not entirely surprising. However, it did raise a significant unmet need to allow for efficient and precise data extraction that does not require direct input from patients or caregivers. Figure 2C shows new generative AI-based tools are able to interpret genetic testing results for patients, caregivers, and physicians.6,7 These kinds of innovations directly bridge the knowledge gap and allow more patients with cancer to get access to the most relevant care options and actively engage in their care decisions. Importantly, this may be applied for any digital system that requires the interpretation of genetic testing reports.
Overall, this study highlights the persistent ethnic disparities in cancer care and the power of digital solutions to bridge these gaps. By integrating AI tools that help patients and caregivers input precise information about their disease, these digital platforms take a significant step toward leveling the playing field in cancer care.
AI-powered analysis precisely identifies underrepresented populations and regions with restricted CT access, paving the way for targeted strategies such as tailored education and support initiatives to promote informed decision-making. The strategic establishment of trial sites in areas with potential matched patients addresses difficulties around travel due to social determinants of health. Ultimately, the use of AI technologies offers a promising and seamless approach to mitigate disparities, improve equitable access to clinical trials, and enhance health care outcomes for underrepresented populations. Through a concerted effort to comprehend and address these disparities, the path to a future characterized by fair and personalized cancer care accessible to all becomes more clearly illuminated.
About the Authors
Avital Gaziel, PhD, is the cofounder and chief science officer at Leal Health in Tel Aviv, Israel.
Yelena Lapidot, PhD, is the medical director, AI domain expert, at Leal Health in Tel Aviv, Israel.