Lung Cancer: Artificial intelligence and machine learning-powered RWE engines are facilitating the development of smarter tools for clinical care.
Lung Cancer Epidemiology Model: This model revealed that the incidence of lung cancer ranges between 6% and 6.9% of all new cancer cases.
By Akansh Khurana
Lung cancer is one of the most commonly diagnosed cancers worldwide and accounts for more deaths than any other type of cancer. In India, smoking and pollution are important contributing factors to lung cancer in addition to environmental toxins, radiations, and gene mutations. Late-stage diagnosis and poor access to affordable healthcare have turned lung cancer into a high-mortality epidemic. Patient-centric care in oncology has been hampered by variations in cancer care and decentralized information systems. Powerful learning systems are required to utilize clinical data, measure the quality of care, provide clinical decision support, and enable learning from every patient. As clinical data infrastructure develops and expands, real-world data (RWD) is set to improve personalized and predictive care in routine clinical practice.
RWD refers to clinical data from electronic health records and longitudinal patient experiences. This is updated on a day-to-day basis with details on molecular profiles, treatment response, outcomes, overall survival, quality of life, and adverse events. RWD can be converted into real-world evidence (RWE) through careful analysis and interpretation and can be used in informed healthcare decision making. THB, India’s leading clinical intelligence company, is currently working on new-age technologies to transform the healthcare industry from gut-based decisions to data and evidence-based decisions and treatment. THB recently developed a Lung Cancer Epidemiology Model to support medical practitioners with algorithms and clinical risk calculators. It maps complete patient journey in lung cancer and includes key touchpoints, gaps, and influencers.
THB’s Lung Cancer Epidemiology Model
This model revealed that the incidence of lung cancer ranges between 6% and 7.2% of all new cancer cases. It also indicated that around 20% of patients receive an inaccurate diagnosis followed by a non-specific treatment for close to 120 days. Post diagnosis, the typical journey of a lung cancer patient is complex, emotionally draining, and full of uncertainty. The patient/family needs to coordinate with a team of doctors – medical and surgical oncologist, radiation oncologist, interventional radiologists, interventional oncologists, pulmonologists, chest and respiratory specialists, gastroenterologists, nutritionists, etc. They may also need to collaborate with other specialists in case of comorbidities like diabetes, hypertension, kidney and cardiovascular diseases.
Several factors were identified that make cancer care challenging in India: a shortage of trained oncologists, poor infrastructure, challenges to access, affordability, lack of awareness in tier 2 cities and rural areas of interior regions, emotional trauma for family, and a delay between the time of symptoms and clinical diagnosis. When non-smokers present with a lung mass, cancer is not considered as a primary/probable condition. Co-morbidities such as coronary artery disease, cerebrovascular diseases, tuberculosis, diabetes, and chronic obstructive pulmonary disease further complicate cancer management and significantly impact the person’s quality of life.
Lung cancer cases are increasing among the young population and age of presentation has gone down to mid-thirties and forties. Most of the cases are advanced-stage carcinomas and highly metastatic. Lung cancer treatment in India is slowly moving to the target-based approach from the traditional histology- based approach. Therapies such as nivolumab and pembrolizumab have been launched in India during 2016 and 2017 but acceptance is significantly low due to the cost and lack of evidence on the Indian population. As of now, only 1/1,000 advanced-stage NSCLC patients can afford new therapies. Affordability issues lead to treatment delays. Delay in accurate diagnosis is likely to worsen prognosis, often to next stage where surgery is not a viable option.
Mapping patient’s journey with a clinical decision support system
A clinical decision support system helps clinicians to make informed decisions related to prognoses, risk stratification, treatments, patient-reported outcomes, and much more. Evidence collected from treatments and clinical observations can help to identify patients with unmet needs and reduce uncertainty in clinical decision making. Artificial intelligence and machine learning-powered RWE engines are facilitating the development of smarter tools for clinical care. In recent news, a Google algorithm out-performed six radiologists in lung cancer screening. This algorithm was trained on data from 42,000 patient CT scans. It detected 5% more cancers and reduced false positives by 11%. These new technologies can speed up screening and support clinicians and radiologist to make better decisions.
RWE can also be used to support public health and epidemiology data: It can be utilized to identify non-malignant (social and geographical) factors that influence diagnosis, screening, treatment, and survival in cancer.
RWE in precision medicine, clinical trials, and post-market settings
Lung cancer treatment is now increasingly biomarker-driven and precision medicine is used to characterize, identify, and target tumor. RWE can fuel precision medicine by studying specific/special populations at large, assessing new therapies, and knowing more about rare diseases with limited evidence from randomized clinical trials (RCTs). RWE can add value to outdated clinical trial models in which only specific people are studied based on
inclusion/exclusion criteria. RWE enables evaluation of patient data in a real-world environment across the region.
With RWE, efficacy and safety of therapy can be assessed in real-time, beyond the periods measured in RCTs. Newly introduced therapies can be compared to competitor molecules in actual patients. The US FDA encourages the use of RWE to support regulatory decision making, including approval of new indications for approved drugs. Pharmaceutical companies can leverage this opportunity to expand indications, support safety reports, and identify new adverse drug reactions in existing therapies.
(The author is co-founder and the CEO of THB – Technology | Healthcare | Big Data Analytics. Views expressed are personal.)