Menu
More than six people die every hour in the US from a blood cancer. Solutions can’t come fast enough for those who suffer with these cancers all around the world. Fortunately, researchers studying blood diseases have experienced rapid advances in their capabilities to develop and test effective therapies with some extremely significant advancements.
Some of the most difficult limitations of molecular profiling for hematological cancer disorders include accurate detection of mutations in GC-rich gene regions and insertions or deletions in challenging genes. Data analysis on NGS DNA samples identifies complex variants to accurately identify myeloid malignancies. This validation of targeted mutations has encouraged many medical centers to order NGS testing for every acute myeloid leukemia case.
Faster, more efficient NGS analysis can drive better hematological cancer research outcomes to potentially improve care for patients with blood cancers and diagnosis of new cases.
International guidelines for hematological cancer diagnosis and treatment are continuously evolving and create the need for laboratories’ fast adaptation. Those evidence-based guidelines by physician commissions contribute to improving the clinical standard of care. The World Health Organization, European Hematology Association, European LeukemiaNet, College of American Pathologists and the American Society of Hematology call for increased use of NGS testing for initial diagnostic workup of blood cancers.
Detection of the relevant biomarkers for myeloid malignancies by NGS, per international guidelines, helps to ensure optimal clinical trial enrollment, therapy validation, dose protocols and other research benefits. A solution that can be constantly updated and inform based on those guidelines ensures that the research is always current.
The accurate assessment of biomarkers and the validity of resulting research findings depend on reliable DNA and RNA fusion panels and easily reproducible results. Data analysis and reporting in a comprehensive platform eliminates silos of valuable data and maximizes its application.
The SOPHiA DDM™ Platform enables the upload of multimodal data from any environment to one of the world’s largest networks of connected labs. Data remains the property of the healthcare institution, but pseudonymized and pooled with like data, it can propel research and ultimately treatment forward with the goal of improved patient care.
Learn more about the capabilities of SOPHiA DDM™ Platform for myeloid biomarker detection and more by contacting us today.
Predictive analytics refers to using statistical algorithms, machine learning techniques, and historical data to forecast future events. In clinical trials, this means integrating diverse data sources – such as, past clinical trials, patient records, and real-world evidence (RWE) – to provide more accurate predictions about trial outcomes, patient responses, and potential risks3. By analyzing patterns from this data, predictive analytics offers a powerful tool for improving the efficiency, accuracy, and safety of clinical trials1.
Traditionally, clinical trials have been labor-intensive and costly, often taking years to yield results4. Researchers had to rely on historical outcomes, guesswork, and incomplete data to design trials and predict success. Predictive analytics changes this paradigm by enabling data-driven decision-making5. By analyzing data from clinical trials, and real-world data (RWD) – including but not limited to patient demographics, electronic health records (EHRs), and claims data-, predictive models can help physicians and researchers make informed decisions about trial design, patient selection, and potential treatment outcomes.
Predictive analytics is particularly valuable because it can integrate multiple data modalities, such as clinical, biological, genomic, biomarker, and imaging data. The ability to combine and analyze this wealth of information is central to predictive analytics’ potential to revolutionize the clinical trial process.
As the healthcare landscape continues to evolve, predictive analytics will play an increasingly central role in the clinical trial process, driving innovation and improving patient outcomes6.
As with any emerging technology, the adoption of predictive analytics in clinical trials requires collaboration between pharmaceutical companies, regulatory agencies, and healthcare providers. However, the potential benefits are too great to ignore. With predictive analytics at the helm, the future of clinical trials looks promising, offering a path to faster, safer, and more effective drug development.
By leveraging on machine learning, SOPHiA DDM™ facilitates the integration and standardization of diverse data modalities – including but not limited to clinical, biological, radiomics, genomics, and digital pathology data – generating powerful insights to support you in accelerating drug development.
Our multimodal AI data analytics helps you optimize your clinical trial and enhance your post-launch access strategy, by predicting patient response to treatment, disease progression, risk of developing adverse events, and supporting treatment decision-making.
A great example of how SOPHiA GENETICS is spearheading innovation in cancer research by applying predictive analytics is the collaboration with UroCCR, the French Kidney Cancer Research Network, to develop a multimodal machine-learning model for predicting post-operative outcomes for individuals facing renal cell carcinoma (RCC). Using real-world prospective data from the UroCCR network, one of the world’s largest collaborative kidney cancer databases, this study showed that the AI model co-constructed by SOPHiA GENETICS and UroCCR provided a strong prediction for postoperative outcomes, outperforming the predictive performance of most usual risk scores. The results of this study have recently been published in npj Precision Oncology.
With a global network of 780+ institutions, across 70+ countries, and over 1.8 million genomics profiles analyzed to date, the SOPHiA DDM™ Platform accelerates the advances in the field of precision medicine. To learn more about SOPHiA DDM™ for Multimodal and our flagship programs visit our page.
The posters provided real-world data insights on the genomic landscape of specific biomarkers associated with lung and prostate cancers, in a subset of European countries.
The first poster spotlight features Stefano Cheloni, Senior Bioinformatician - Tertiary Analysis at SOPHiA GENETICS, presenting the poster titled: “Real-world insights from France, Italy, Spain, and Austria for the investigation of common (exons 19,21) and rare (exons 18, 20) EGFR mutations in lung cancer”.
The identification of exon-specific EGFR mutations can guide the appropriate treatment of lung cancer with tyrosine kinase inhibitors (TKIs) or alternative targeted therapies. This project aimed to explore the landscape of next-generation sequencing (NGS) testing practices for EGFR mutations in clinical practice in a subset of European countries, to determine the potential number of individuals with lung cancer that could benefit from treatment with TKIs or alternative targeted therapies.
Read the EGFR poster through here.
Watch the spotlight below:
In our second poster spotlight, Adrian Janiszewski, Manager Bioinformatician - Team Lead Tertiary Analysis at SOPHiA GENETICS, showcases a poster titled: “NGS testing practices and molecular profiles of BRCA1/2 in prostate cancer: Real-world insights from France, Italy, Spain, and Austria”.
BRCA1/2 testing can inform which prostate cancer patients might respond to PARP inhibitors (PARPi). Having a better understanding of BRCA1/2 testing practices in the real-world setting can provide valuable insights into identifying gaps and opportunities to improve the identification of metastatic prostate cancer (mPC) patients who may benefit from PARPi treatment. In this study we investigated NGS testing practices results across specific European countries.
Read the BRCA 1 & 2 poster through here.
Watch the spotlight below:
We would like to warmly thank Stefano and Adrian for their insightful presentations and for sharing the key takeaways of these projects.
Learn more about SOPHiA DDM™ for BioPharma, by visiting the page.
DISCLAIMER:
All product and company names are trademarks™ or registered trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them. SOPHiA GENETICS products are for Research Use Only and not for use in diagnostic procedures unless specified otherwise.
Explore this infographic summary to learn more about the machine learning model developed by Margue et al. for the prediction of disease-free survival in patients undergoing surgery for renal cell carcinoma.
Margue G, et al. UroPredict: Machine learning model on real-world data for prediction of kidney cancer recurrence (UroCCR-120). NPJ Precis Oncol. 2024 Feb 23;8(1):45.
SOPHiA GENETICS multimodal analysis technology and concepts in development. May not be available for sale.
Multimodal healthcare datasets synergistically integrate diverse data modalities such as genomic, clinical, radiomic, proteomic, and biological data, to provide comprehensive insights into human biology and medical conditions. Multimodal datasets have the potential to predict outcomes more accurately and informatively than the sum of their parts (Fig. 1).
Figure 1.Multimodal healthcare data integrated and analyzed by artificial intelligence (AI)/machine learning can provide useful information for healthcare professionals to use to improve patient care.
Genomics data.
Radiomics data include x-rays, CT scans, MRI scans, ultrasound images, and mammograms.
Clinical and biological data from electronic health records include patient histories, demographics, notes, diagnosis codes, procedure codes, laboratory results, and vital signs.
Proteomics data.
Digital pathology data.
Patient-reported data includes questionnaires and health journals, as well as data from wearable devices monitoring heart rate, sleep patterns, and activity levels, and implantable devices such as pacemakers, insulin pumps, and continuous blood glucose monitors.
Environmental data includes air quality and location data.
New data-driven technologies powered by novel ways of linking and analyzing patient data are set to transform the way that healthcare is delivered.1 Healthcare professionals routinely make use of multiple sources of data to arrive at a diagnosis and to decide on patient management.2 However, a significant level of expertise is required for an in-depth understanding of even a single data type (e.g. radiological images) such that it is unfeasible for individual healthcare professionals to master all areas. AI/machine learning technologies can be leveraged to bring together and analyze multimodal healthcare data, breaking data silos and creating robust and accurate predictive models.3 With the appropriate guidance around decision-making and communication, the valuable insights gained from these predictive models have the potential to support healthcare professionals to improve patient care.
Machine learning technologies can integrate data from disparate multimodal sources to provide a holistic understanding of patients’ health and medical conditions. Data are combined from multiple modalities with the aim of extracting complementary information to power predictive models that can find relationships between different variables/features that are not clearly visible or known by healthcare professionals. Indeed, multimodal data fusion models have consistently shown to provide increased accuracy (1.2-27.7% higher) and performance (AUC 0.02-0.16 higher) than models that utilize data from single modalities for the same task.4
Oncology is one of the medical specialties that most commonly leverages multimodal methods for clinical decision support.5 Machine learning technologies have the potential to explore complex and diverse data to support healthcare professionals from screening to treatment (including relapse).6 Identification of risk factors can support non-invasive patient screening and preventive care.3 Detection of patterns in easily accessible data can help identify diagnostic or prognostic biomarkers to improve patient risk stratification or selection for clinical trials. Identification of predictive signatures of risk factors, adverse treatment reactions, treatment responses, or treatment benefit, can guide decisions around patient management.
Figure 2. The number of PubMed articles published on multimodal oncology data has dramatically increased in recent years.
PubMed search for ((multimodal) AND (oncology)) OR ((multimodal) AND (cancer)).
*2023 analysis includes data available at time of writing (January-September).
With data privacy and security paramount, multimodal healthcare data can also be leveraged to accelerate advances in medical research, such as the discovery of novel biomarkers and therapeutic targets for drug development, as well as supporting population health management by providing a comprehensive view of health trends and outcomes. The rapid increase in peer-reviewed publications on the topic over the last 13 years demonstrates that the extraordinary value of multimodal oncology data is already recognized by the scientific and medical communities (Fig. 2). Leveraging machine learning to collate and analyze the vast diversity of multimodal data for data-driven precision medicine is on track to drive the next revolution in healthcare.
SOPHiA DDM™ multimodal healthcare analytics will have the potential to break data silos by streamlining the integration of longitudinal oncology data from multiple sources and modalities – including but not limited to genomic, radiomic, digital pathology, biological, and clinical data. The SOPHiA DDM™ Platform uses machine learning-powered analytics to assemble, standardize, and transform multimodal data into accessible data-driven insights, facilitating the identification of multimodal predictive signatures, as well as treatment response patterns and trends. To learn more and get in touch, visit the webpage.
Product in development – Technology and concepts in development. May not be available for sale.
Area under the ROC curve (AUC) – A ROC (receiver operating characteristic) curve is a graph that plots true and false positive rates to demonstrate the performance of a model. AUC measures the area underneath the ROC curve to provide an aggregate measure of performance. AUC values range between 0 and 1, with a score of 0 meaning that all predictions are wrong, and a score of 1 meaning that all predictions are 100% correct. Essentially, AUC represents the probability that a positive result is truly positive and a negative result is truly negative.
Omics data – Large-scale information related to the biology of organisms.
Digital pathology images – Scanned images of tissue samples on glass slides.
Companion diagnostics (CDx) are medical devices, specifically an in vitro diagnostic device (IVD), providing important information regarding the safe and effective use of therapeutics. The Food and Drug Administration (FDA) ascribes them three crucial functions: 1) identify patients more likely to benefit from a therapeutic; 2) determine patients at increased risk of serious side effects; and 3) monitor treatment responses for the purpose of adjusting dosage or regimens to improve safety and effectiveness. In our estimation, CDx act as a compass that directs the healthcare provider to the most appropriate treatment for each patient1.
The inception of CDx can be traced back to 1998, when the FDA granted concurrent approval for trastuzumab, a targeted cancer drug, and HercepTest™, a HER2 immunohistochemical assay. This milestone marked the birth of the drug-diagnostic co-development model, a transformative approach that has since witnessed consistent and substantial adoption2.
However, over the next 14 years, CDx advancement was slow, with the majority of new approvals occurring only in the past decade. In fact, from 1998 to 2012, approximately 20 new CDx were approved, whereas from 2013 to 2023, that number rose to 1343.
Today, approximately 50% of all new molecular entity (NME) approvals in oncology have an associated CDx or biomarker listed in the label required for safe and effective use (based on FDA approvals from 2021 and 2022 of NME in oncology3). Despite the historical tendency toward oncology products, applications in rare diseases and metabolic syndromes are evolving, paving the way for CDx to become an intrinsic part of precision medicine clinical trials across many indications.
The use of a companion diagnostics strategy in clinical trials, which we define here as using one or more biomarkers to pre-select and enroll patients more likely to respond to the experimental therapy, is commonly employed in oncology. In these trials, identification and pre-selection have significant advantages, allowing smaller patient groups to power the statistical analysis, potentially reducing overall costs, and increasing the likelihood of approval4.
But while a CDx strategy makes regulatory approval of a cancer drug more likely, it can simultaneously add complexity to the process:
Despite the evident complexity, the widespread adoption of next-generation sequencing (NGS) has made using companion diagnostics and deploying biomarker-driven strategies in clinical trials easier by permiting screening for multiple biomarkers simultaneously. Rather than rely on a one-biomarker-one-test model, NGS permits patients to be screened for eligibility for multiple therapeutics or clinical trials.. Moreover, targeted gene panels and broader approaches, such as comprehensive genomic profiling (CGP) and whole exome sequencing (WES) have further streamlined the development of multi-biomarker-driven CDx.
Returning to our analogy, while CDx acts as the compass, NGS technologies are the roadmap to determine the most suitable treatment for each patient.
There are two well-defined pathways to approval for companion diagnostics:
HercepTest™ followed the first pathway where the physical kit, produced by an IVD manufacturer, received a Pre-market Approval (PMA). In contrast, the precedent for the single-site model was only established much later with the publication of clinical evidence supporting the hypothesis that BRCA-mutated patients were more likely to benefit from treatment with olaparib—a PARP inhibitor first approved for the treatment of advanced ovarian cancer5,6.
Requiring a complex workflow and expert oversight, BRCA analysis was more conducive to the simplicity of the single-site model since the very nature of this pathway streamlines validation – validating a workflow within a single lab is less time-consuming than across multiple labs. This new approach played a pivotal role in rapidly advancing the commercial adoption of NGS applications.
Today, most NGS-powered CDx assays follow the single-site pathway3, creating a new set of challenges. Despite the increased simplicity, this model confines assays to single locations, limiting the capacity for sample analysis, significantly increasing turnaround times, and reducing patient access. In the new age of precision medicine, these limitations are being addressed through a decentralized testing and analysis model supported by technology-agnostic and easy-to-implement workflows.
While direct co-development of a CDx and therapeutic stands as the preferred regulatory model by the FDA, alternative approaches may be utilized due to the inherent challenges of aligning IVD and drug development.
While the development of a CDx and therapeutic are tightly entwined, drug developers and IVD manufacturers remain separate entities with a few exceptions. This requires the carefully selection of partner(s) within the IVD and CDx ecosystem to ensure successful programs.
Many questions must be addressed early in the process, as even suboptimal approaches can significantly delay and impact commercial uptake:
The advent of NGS technologies has heralded a healthcare revolution, propelling us toward data-driven precision medicine. Yet, as we push ahead in developing biomarker-driven applications for a plethora of indications, we face the potential for increased implementation challenges, threatening to complicate the patient journey.
The adoption of a decentralized, globally accessible, intuitive, and technology-agnostic SOPHiA DDM™ Platform, leveraging proprietary algorithms and a vast portfolio of robust NGS-based applications, is well positioned to streamline CDx co-development and implementation, enhancing access to analytically robust solutions without overtaxing healthcare resources.
Our holistic approach to co-development is poised to chart a course toward a more integrated future, arming developers with the necessary data and insights to tackle potential hurdles and maximize the time and resources allocated to clinical research programs.
At SOPHiA GENETICS, our unwavering commitment is to collaboratively engineer deployable solutions that elevate implementation and accessibility in precision testing while streamlining the process of analysis and interpretation. Explore the possibilities of SOPHiA DDM™ for BioPharma by visiting our dedicated page.
References
1 Food and Drug Administration (FDA). In Vitro Companion Diagnostic Devices: Guidance for Industry and Food and Drug Administration Staff. Issued on: August, 2014. Accessed on: October, 2023. Retrieved from: https://www.fda.gov/media/81309/download
2 Jørgensen JT. Drug-diagnostics co-development in oncology. Front Oncol. 2014;4:208. doi: 10.3389/fonc.2014.00208
3 FDA. List of Cleared or Approved Companion Diagnostic Devices (In Vitro and Imaging Tools). Accessed on: October 2023. Retrieved from: https://www.fda.gov/medical-devices/in-vitro-diagnostics/list-cleared-or-approved-companion-diagnostic-devices-in-vitro-and-imaging-tools
4 Biotechnology Industry Organization (BIO). Clinical Development Success Rates and Contributing Factors 2011–2020. Accessed on: October 2023. Retrieved from: https://go.bio.org/rs/490-EHZ-999/images/ClinicalDevelopmentSuccessRates2011_2020.pdf
5 Ledermann J, et al. Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial. Lancet Oncol. 2014;15(8):852-61. doi: 10.1016/S1470-2045(14)70228-1.
6 Deeks ED. Olaparib: first global approval. Drugs. 2015;75(2):231-240. doi: 10.1007/s40265-015-0345-6
7 Hanna TP, et al. Mortality due to cancer treatment delay: systematic review and meta-analysis. BMJ. 2020 Nov 4;371:m4087. doi: 10.1136/bmj.m4087
Precision medicine, also known as personalized medicine, aims to enhance healthcare quality by tailoring treatments to each person's unique genetic makeup, environment, and lifestyle. While a fully individualized approach to medicine is still a work in progress, the recognition that patient heterogeneity influences treatment effectiveness is not new1.
Historically, medicine has heavily relied on trial-and-error strategies for discovering, developing, and testing new treatments targeted at specific indications. This disease-centered approach resulted in predetermined standard therapies tailored to the “average patient.” While this one-size-fits-all approach has succeeded in many indications, it also carries significant drawbacks, particularly when dealing with complex diseases such as cancer or inherited disorders. In these cases, the risk of adverse side effects (e.g., toxicity) and reduced therapeutic response often result in poorer patient prognoses and quality of life1.
Precision medicine represents a patient-centric paradigm shift, acknowledging each individual's uniqueness while using real-world data and advanced statistics to guide the discovery-to-treatment process. For instance, pharmacogenomics requires us to look at each patient genomic data individually and in the context of others, enabling stratification into cohorts for predicting treatment responses2.
Success in precision medicine hinges on the ability to derive meaningful insights from large patient datasets. Fostering data diversity has the potential to further advance progress in this area1,3.
Recent technological advances have made precision medicine more accessible and impactful than ever before. Next-generation sequencing (NGS) has become more affordable, transforming it from a research-focused technology into a tangible clinical reality. This progress was further propelled by the widespread adoption of electronic health records (EHRs) and laboratory information management systems (LIMS), which not only facilitate population-scale research but also enable the use of clinical decision support tools for the delivery of targeted therapies to individual patients4.
The ability to identify genetic biomarkers and assess variant pathogenicity has grown significantly in the past decade. This has not only revolutionized patient diagnosis but also transformed drug development. A pivotal moment was the approval of imatinib by the FDA in 2001, the first small molecule targeted therapy for chronic myeloid leukemia (CML)5. By inhibiting the BCR-ABL fusion protein, imatinib was shown to effectively slow the progression of CML from chronic phase to blast crisis, making it the first of its kind.
This groundbreaking milestone paved the way for the approval of many other targeted therapies, such as gefitinib targeting EGFR alterations associated with NSCLC (2003) and trastuzumab for HER2-positive breast cancer (2006). The pace of new targeted drug approvals continues to accelerate year after year, heralding a promising era of precision medicine6.
Timeline of FDA-approved targeted therapies in cancer. Grey bubbles represent the relative number of approvals per year. Data source: Waarts et al 2022.
To achieve a truly personalized approach to medicine, the harmonization of translational and precision medicine is paramount. This coordination between early mechanism-based drug development and late-stage patient-centric approaches gives rise to an end-to-end biomarker-guided process, allowing us to optimize treatment strategies for patient cohorts right from the outset7.
Known as translational precision medicine, this emerging concept brings a fresh perspective to the translational gap, calling for a broader scope beyond a purely genetic-based definition of biomarkers and introducing a multimodal approach by taking into account a wider range of healthcare variables. To make this new concept a reality, significant technological progress is required in several key areas8:
By addressing these critical areas of advancement, we can pave the way for a future where each patient receives personalized treatments tailored to her or his unique needs and characteristics. The pursuit of translational precision medicine promises to revolutionize healthcare, offering improved patient outcomes and transforming the landscape of medical research and development.
Powered by proprietary algorithms and enriched with data from 750+ institutions*, the SOPHiA DDM™ Platform accelerates advances in the field of precision medicine. Its core mission centers on empowering clinical researchers across healthcare and biopharma spheres alike.
To learn more about SOPHiA DDM™ BioPharma Solutions for biomarker-centric discovery, development, and application deployment, visit our page.
* The number of institutions represents active customers who have generated revenue through the SOPHiA DDM™ Platform usage or Alamut™ Visual Plus licenses as of September 30, 2022.
An endemic problem in the healthcare industry is that there are too few staff 1 to give all patients the time that they need to receive the best care
Precision medicine is set to revolutionize healthcare,2 and state-of-the-art technologies are essential to achieve this. These technologies will be used to detect patterns in large quantities of genetic, biological, clinical (research), and environmental data, to extract insights related to personalized patient care. Furthermore, an endemic problem in the healthcare industry is that there are too few staff1 to give all patients the time that they need to receive the best care. State-of-the-art technologies have the potential to streamline processes so that healthcare professionals (HCPs) have more time to dedicate to their patients.
When discussing state-of-the-art technologies, we inevitably come across terms like artificial intelligence (AI), machine learning (ML), deep learning (DL), and neural networks (NN). And although we might have a vague understanding of what each of these terms means, it is difficult to distinguish between them and to know how to speak about them when discussing their potential impact on healthcare with colleagues and collaborators. This guide will clarify these terms.
Artificial intelligence, machine learning, deep learning, or neural networks?
- Artificial intelligence
- Machine learning
- Deep learning
- Neural networks
How are artificial intelligence and machine learning used in healthcare?
- Predictive medicine and imaging
- Patient support
- Health services management
- Physical assistance
- Drug development
How will AI and machine learning impact the future of healthcare?
Conclusion
In a nutshell, artificial intelligence (AI) is when computer systems simulate human intelligence by performing tasks that typically require human cognitive abilities.
Machine learning (ML) is one of the key techniques used in AI, where algorithms and statistical models are designed by humans to enable computer systems to learn and improve from experience without being explicitly programmed.
Deep learning (DL) is a subset of ML where algorithms use artificial neural networks to process and analyze large amounts of data, extracting relevant features and patterns.
Neural networks (NN) are a fundamental component of DL algorithms, which are designed to simulate the behavior of the human brain with a network of interconnected nodes (or neurons) that process and transmit information.
For more detailed explanations, click the terms below to reveal.
AI is a field of computer science that aims to create intelligent machines that can perform tasks such as visual perception, speech recognition, decision-making, and natural language understanding. AI systems use algorithms and statistical models to process large amounts of data, identify patterns, and make predictions or recommendations based on the data.
Although the term AI was coined in 1956 by John McCarthy, the possibility that machines could simulate human behavior and “think” was raised earlier by Alan Turing in 1950.3 Since then, computational power has grown exponentially, and AI is integrated into our daily lives in many forms. For example, many of us use (the likes of) Siri, Alexa, or Google Assistant without a second thought.
More recently, AI has become part of medical practice, where it can improve patient care by speeding up processes and achieving greater accuracy.3 Radiological images, pathology slides, genetic information, and patients’ electronic medical records can be evaluated using AI to aid with administrative tasks and diagnosis and treatment decisions, enhancing the capabilities of clinicians.
ML is an AI technique for fitting models to data that involves the development of algorithms and statistical models.4 The machine “learns” by training models with data to identify patterns so that it can make predictions or decisions. Machine learning is a widely used AI technique and forms the basis of many approaches within the field.
In healthcare, the most common application of ML is in precision medicine, where patient data are used to predict which treatment strategies are most likely to succeed.4 In order to make predictions, the algorithms generally require a training dataset for which the outcome variable (e.g., onset of disease) is known – this is called supervised learning.
The most complex forms of machine learning involve DL, or NN models, which have many layers of features or variables that predict outcomes.4 The more layers a network has, the deeper it is, hence the term “deep learning”. The improved capabilities of today’s graphics processing units and cloud architectures make it possible to process and analyze thousands of hidden layers of features.
In healthcare, pattern recognition through DL involves teaching a computer what certain groups of symptoms or radiological images, for example, look like via repetitive algorithms.5 The algorithms enable the computer to learn and improve from experience by adjusting weights and biases. An example of this is when Google’s artificial brain project trained itself to recognize cats based on 10 million YouTube videos, with recognition efficiency improving the more images it reviewed.
DL algorithms can be used for a wide range of applications, such as image and speech recognition, natural language processing, and autonomous systems. A common application of DL in healthcare is for the identification of clinically relevant features (e.g., tumors) in imaging data, which may not be perceived by the human eye.4 Using DL to analyze radiology images can provide a more accurate diagnosis than the previous generation of automated tools for image analysis, computer-aided detection (CAD).
NNs are a fundamental building block of DL algorithms, which can learn and make decisions by themselves.6 They are an interconnected network of nodes (neurons) that mimic the human brain, with weighted communication channels between them.7 Each neuron receives input from other neurons in the previous layer, applies a mathematical operation to that input, and then passes the output to neurons in the next layer. One neuron can react to multiple stimuli from neighboring neurons and use weights and biases to adjust the strength of connections between them. The whole network can change its state based on different inputs received from the environment. As a result, NNs can generate outputs in response to environmental input stimuli, just like the human brain reacts to the environment around us.
AI and ML can be used in healthcare to assist HCPs in streamlining processes, reducing costs, and perhaps, most importantly, making faster, data-driven clinical decisions, all with the aim of improving patient outcomes. Advances in big data analytics using AI techniques are unlocking clinically relevant information hidden in increasingly available healthcare data, which is successfully assisting HCPs with clinical decisions.8 There are five key areas in which AI and ML are currently accelerating healthcare – predictive medicine and imaging, patient support, services management, physical assistance, and drug development.
ML has the potential to analyze individual patient data to predict risk, support diagnosis, predict disease progression and prognosis, and to identify the most appropriate treatment regimens.9 In addition, ML has the potential to identify risk factors and drivers for each patient, to help target healthcare interventions for better outcomes.
ML has primarily been used in healthcare to analyze data from imaging, genetic testing, and electrodiagnosis.8 These data are analyzed by AI technologies to cluster patient traits and associate them with a diagnosis, or to predict disease outcomes or response to treatment.
The application of ML to medical imaging has been found to improve accuracy, consistency, and efficiency. In 2017, Arterys developed the first US FDA-approved clinical cloud-based DL application, CardioAI.6 CardioAI analyzes cardiac magnetic resonance images (MRIs) to provide information such as cardiac ejection fraction in a matter of seconds, and has since expanded to cover additional organs and imaging techniques. The time-saving implications from introducing AI support platforms into clinical practice can be quite substantial; radiologists can save ~1 hour per day interpreting chest CTs,10 DL can measure pediatric leg lengths 96x faster than subspecialty-trained pediatric radiologists,11 and AI systems can automate the triaging of adult chest radiographs.12
ML has successfully been used to screen for diabetic retinopathy, identify nonmelanoma and melanoma skin cancers, predict seizures, predict bladder volume, predict cardiovascular risk, and predict progression of Alzheimer’s disease and response to drug therapy.6,7
There are multiple ways in which AI has been used to provide outpatient care. In combination with robotics, AI has been harnessed to restore movement control in patients with quadriplegia for example, and to control prostheses.8 Rehabilitation robots can physically support and guide a patient's limb(s) during physical therapy.9 AI can also be used to assist the independent living of elderly and disabled people with tools such as fall detection systems and wheelchairs controlled by facial expressions.7
AI-powered virtual assistance like chatbots and voice assistants can provide patients with personalized medical advice and support from home, helping patients to manage their own health more effectively with the aim of reducing the workload of HCPs.4,5 Furthermore, wearable systems can support continuous patient monitoring and healthcare delivery.9
AI can support HCPs to work more efficiently, freeing up more time to spend on patient care. AI systems can provide HCPs with real-time medical information updates, coordinate information tools for patients, optimize logistics processes, benchmark data for analyzing services delivered, and much more.9 Process automation, specifically, can be leveraged for tasks such as claims processing, clinical documentation, revenue cycle management, and medical record management.4 Ultimately, the administrative assistance provided by AI creates more time for human interactions.
AI has the potential to transform surgical robotics through devices that can perform semi-automated surgical tasks with increasing efficiency.9 AI technologies can guide surgical tools and make more precise movements than possible within our capacity as humans.5
AI techniques streamline the design and development of new drugs by analyzing the vast amount of data available from clinical trials and databases to identify new drug targets and predict drug efficacy and safety.9
AI and ML are set to drive the future of healthcare. In particular, ML is a key component in the advancement of precision medicine.4 AI and ML will greatly enhance risk prediction and diagnosis of diseases, and will facilitate personalized treatment strategies based on a broad spectrum of individual patient characteristics. It seems feasible that most radiology and pathology images will soon be examined by a machine and that AI and ML will help HCPs to remotely monitor patients. Speech and text recognition are already used for patient communication and to capture clinical notes, and their usage is likely to increase. AI and ML will also help to accelerate and reduce costs associated with the drug development process.
The successful integration of AI and ML technologies into healthcare requires more than just reliability and accuracy. Several critical factors must be in place to ensure sustainable adoption, including integration with electronic health record systems, standardization, adequate funding, improved regulatory approval processes, staff training, and continuous algorithm optimization with new data.
However, the most crucial factor is transparency. The complexity of AI/ML algorithms and models can make them difficult to interpret or explain, potentially raising concerns about accountability, trust, and privacy. To promote the responsible and sustainable adoption of these technologies, healthcare institutions and regulatory bodies must first establish governance mechanisms and monitoring structures to safeguard the interests of providers and patients alike. This will not only ensure a smooth transition but also foster trust in the use of AI technologies in healthcare.
It seems increasingly clear that AI systems will not replace human HCPs but will instead enhance their capabilities to improve the care of patients. It is important that HCPs are trained and provided with the skills to efficiently work alongside AI. The goal will be to balance a mutually beneficial relationship that leverages on the speed and analytic potential of AI with the uniquely human strengths of empathy, nuance, and seeing the big picture.
Whether AI, ML, DL, or NN, the overall field of artificial intelligence is booming, and the possibilities for improving healthcare are exciting. Although it may seem fantastical, HCPs are already working alongside machines to enhance clinical decision-making, and the future in this field has huge potential for improving patient care.
At SOPHiA GENETICS, we have a strong background in developing ML algorithms that aim to extract actionable insights from genomic data. SOPHiA DDM™ for Multimodal (product in development) leverages on propriety ML algorithms, with the potential to maximize the value of multimodal health data. Our technology has so far been utilized in retrospective research studies to assess prognosis and predictive biomarkers in non-small cell lung cancer (NSCLC),13 investigate new meningioma biomarkers to further understand treatment response patterns,14 evaluate pathological complete response status and treatment response in patients with early triple negative breast cancer (TNBC),15 and estimate the risk of disease upstaging, disease-free and overall survival in kidney cancer.16
These studies report data associated with products or concepts in development. They are not available for sale and not intended for use in diagnostic procedures or treatment decisions.
To learn more about SOPHiA DDM™️ Multimodal Healthcare Analytics for the visualization of longitudinal patient data and cohorting, explore here or request a demo here.
At ESMO 2022, oncology experts gathered in Paris and online to share and debate the new developments in the field of medical oncology. This year's program featured more than 20 tracks covering all tumor types, therapeutic innovations, translational research, patient advocacy, public policy, and more...
Discover our summary of three compelling talks showcasing the journey towards precision therapy in various tumor types.
Genomic profiling and molecular targeting of lung cancer brain metastases1
Haiying Cheng, Dept. Medical Oncology, Montefiore Medical Center/Albert Einstein College of Medicine, Bronx, NY, USA
Approximately 57% of patients with non–small-cell lung cancer (NSCLC) present with metastatic disease2. Among them, brain metastases (BM) affect up to 45% of all cancer patients and arise from lung cancers in 40-50% of the cases3. There have been limited studies investigating the genetic signatures of LC BM, and with small cohorts so far.
Assembling a large number of lung cancer cases (47215 NSCLC; 29438 lung adenocarcinoma), Dr Cheng and colleagues looked for key genetic alterations in loco-regional lesions (Loco), extracranial metastases (EM), and BM with comprehensive genomic profiling (CGP). They found significantly more genetic alterations in the PI3K/AKT/mTOR pathway in BM (Loco 13.0% vs EM 14.5% vs BM 18.1%), primarily driven by RICTOR amplification (Loco 3.6% vs EM 6.2% vs BM 8.6%).
RICTOR amplification is the most enriched actionable genomic target in NSCLC brain metastases.
Furthermore, in vitro genetic knockdown and pharmacological inhibition of RICTOR significantly reduced migration and invasion in RICTOR-amplified NSCLC cells, whereas RICTOR upregulation promoted these processes, modulating the AKT, MET, EMT, and CXCL12 chemokine-CXCR4 pathways. Finally, in vivo studies in orthotopic mouse models revealed that both RICTOR and mTOR1/2 inhibition significantly reduced lung cancer tumor growth and spread in the brain.
Dr Cheng provided evidence for the benefit of further investigation on the development of RICTOR-targeted therapeutic strategies for the treatment and/or prevention of lung cancer BM. This study is a good example of how genomic profiling, combined with functional analyses, can identify new potential therapeutic targets.
Neoadjuvant immune checkpoint inhibition in locally advanced MMR-deficient colon cancer: The NICHE-2 study4.
Myriam Chalabi, Gastrointestinal Oncology, Netherlands Cancer Institute, Amsterdam, Netherlands
Mismatch repair deficiency (dMMR) is observed in ~15% of colorectal cancers (CC)5 and 1/3 is associated with Lynch Syndrome. This characteristic genetic signature is marked by high levels of microsatellite instability (MSI) and resistance to standard-of-care neoadjuvant chemotherapy (5-7% pathological response (≤50% residual viable tumor; PR)). NICHE-1 exploratory study showed the potential of neoadjuvant immunotherapy in patients with dMMR CC with extraordinary PR in 100% of the patients6.
Dr Chalabi presented the NICHE-2 investigator-initiated study, conducted in 6 hospitals in the Netherlands. 107 patients with non-metastatic untreated dMMR CC and mainly high-risk tumors received injections of nivolumab and ipilimumab within 6 weeks prior surgery. The impressive pathological tumor regression was shown in a waterfall plot that led to a standing ovation! 95% of the treated patients showed PR, and 67% had no residual viable tumor (complete PR; cPR), contrasting with the neoadjuvant chemotherapies in the same patient population. Only 4% experienced grade 3-4 immune-related adverse events and 98% of patients underwent timely surgery, meeting the safety primary endpoint. To date, no disease recurrence has been observed and the 3 years disease-free survival data are expected next year.
Neoadjuvant immunotherapy has the potential to become standard of care for patients with dMMR colon cancer.
NICHE-2 trial opens the possibility that a surveillance approach may be possible for some patients with early dMMR CC and gives a glimpse at the potential of translational research to identify predictive biomarkers in pre- and post-treatment samples. While those preliminary results are extremely promising and we surely wait for the longer-term follow-up data to confirm them, patient selection remains crucial. Indeed, neoadjuvant decisions are based on radiological assessment of the tumor, particularly difficult in dMMR cancers, as well as the biopsy, and Dr Chalabi highlighted the need for improvement in the imaging techniques and circulating DNA analyses.
Final overall survival results from the phase III PAOLA-1/ENGOT-ov25 trial evaluating maintenance olaparib plus bevacizumab in patients with newly diagnosed advanced ovarian cancer7.
Isabelle Ray-Coquard, Department Of Medical Oncology, Centre Léon Bérard, and GINECO, Lyon, France
The late diagnosis of advanced ovarian cancer (AOC) is often accompanied by relapse, despite surgery and platinum-based chemotherapy. Treatment with olaparib (ola), a poly(adenosine diphosphate–ribose) polymerase inhibitor (PARPi), provided progression-free survival (PFS) benefit as maintenance therapy in patients with AOC carrying mutations in BRCA1 or BRCA2 (BRCAm)8. Besides, the incorporation of the antiangiogenic agent bevacizumab (bev) is a recognized option in addition to chemotherapies9.
PAOLA-1 investigators conducted a phase III trial where 806 patients with AOC and after first-line platinum-based chemotherapy plus bev were randomly assigned in a 2:1 ratio to ola + bev or placebo (pbo) + bev treatment. The primary endpoint was the PFS. They showed that combined treatment with ola + bev reduced the risk of relapse by 41% compared to bev alone, reducing by 67% in HRD+ patients (exhibiting BRCAm and/or genomic instability score ≥42)10.
Here, Dr Ray-Coquard presented the final overall survival (OS) results, a key secondary endpoint. She showed that the OS rate after 5 years was not different between the two arms (47.3% vs 41.5%) but significantly increased for HRD+ patients treated with ola + bev (65.5% vs 48.4%), regardless of BRCAm status. Also, PFS was significantly increased in the same population (46.1% vs 19.2%).
Maintenance therapy with olaparib plus bevacizumab improved survival in HRD+ patients with newly diagnosed advanced ovarian cancer.
With the absence of new safety signals and major adverse effects, these data confirmed the benefit of olaparib and bevacizumab combination as a standard of care for HRD+ patients and reinforced the importance of precision medicine and biomarker testing to guide treatment decisions.
At ASCO 2022, oncology professionals gathered in Chicago, Illinois, and online to discuss the latest advances in research and care for patients with cancer. This year's program featured over 200 sessions on Advancing Equitable Cancer Care Through Innovation. The presentations spanned from care delivery and regulatory policy to developmental therapeutics, gastrointestinal cancer, lung cancer, pediatric oncology, and beyond. Here, discover our summary of four outstanding ASCO presentations focusing on breast cancer treatment, diagnosis and follow-up, showcasing the power of precision medicine in healthcare.
Targetable genomic mutations in young women with advanced breast cancer1.
Norin Ansari, Yale New Haven Hospital, New Haven, CT
Advanced breast cancers (BC) in young women (under 40 years old) are often more aggressive and with worse prognoses than in older women. As treatment strategies can be dictated by the type of genomic alteration (GA), knowledge of BC genetic profiles across ages can greatly improve guidance and outcomes. In her poster presentation, Norin Ansari analyzed over 2,000 BC using hybrid-capture based comprehensive genomic profiling (CGP) to evaluate subtypes of GA and confirmed via immunohistochemistry (IHC) hormone receptors (HR) and PD-L1 status.
The study showed a mutations stratification within the population of BC depending on patient's age. Indeed, younger patients had higher rates of BRCA1, BRCA2, and RB1 mutations and lower rates of CDH1 and PIK3CA mutations than did older patients. Differences were statistically significant in BRCA1, CDH1, and PIK3CA. Norin Ansari also showed that breast tumors in younger women were less likely to be estrogen receptor positive (ER+) and more likely to be triple negative while no clear age-related pattern for HER2 status could be highlighted. Finally, younger women were more often PD-L1 positive and had lower tumor mutational burden (TMB) than their older counterparts.
Different mutational profiles may support differential use of targeted and immune therapies.
With increasing availability of targeted and immune therapies, knowing which GA each group of women has allows to better tailor therapies and leads to more effective treatments. For instance, BRCA1 mutations may lend to PARP inhibitor use while PIK3CA mutations may indicate the benefit of alpelisib prescription. The difference in genetic mutations between age groups can give a head start when treating women with breast cancer and CGP can refine the approach for better results.
Alpelisib + fulvestrant in patients with hormone receptor–positive, human epidermal growth factor receptor 2–negative advanced breast cancer: Biomarker analyses by next-generation sequencing from the SOLAR-1 study2.
Dejan Juric, Massachusetts General Hospital Cancer Center, Harvard Medical School, Boston, MA
PIK3CA mutations account for approximately 40% of the hormone receptor positive (HR+), HER2-negative (HER2-) advanced BC. PIK3CA encodes for a subunit of PI3K, key of a highly interconnected pathway regulating growth and cell survival. PI3K pathway alterations are associated with endocrine therapy resistance, hence the poor prognosis for HR+, HER2– advanced BC.
Dejan Juric introduced the SOLAR-1 phase 3 study, a randomized controlled study testing the efficacy of the combined administration of alpelisib (ALP, a PI3Kα-selective inhibitor and degrader) and fulvestrant (FUL, a selective estrogen receptor degrader) in HR+, HER2- advanced BC patients. SOLAR-1 shows improved progression-free survival (PFS) in ALP + FUL treated patients versus placebo + FUL of 11.0 and 5.7 months, respectively. Going one step further, they measured the efficacy outcomes in patients with specific gene alterations (GA) in a PIK3CA-altered cohort, applying a retrospective exploratory biomarker analysis.
SOLAR-1 baseline tumor samples were tested by next-generation sequencing (NGS) and clinical benefit was assessed using PFS and hazard ratio based on TMB and GA status in the PIK3CA-altered cohort. While ALP + FUL clinical benefit was seen across TMB quartiles, it was more pronounced in patients with a low TMB (PFS of 18.5 months with ALP versus 3.2 months with placebo). They also observed that, despite improved PFS with ALP + FUL treatment in all PIK3CA-altered patients, the level of benefit may depend on the mutation status of other genes involved in MAPK pathway, PI3K pathway, in endocrine therapy or CDK4/6 inhibitors resistance. For instance, greater benefit was observed with altered FGFR1/2 but limited in MYC- and RAD21-altered cohorts. Besides, ALP + FUL efficacy was independent of GA in TP53, ESR1, CCND1, MAP3K1 and ARID1A.
Clinical benefit of ALP + FUL was maintained regardless of alterations in most biomarkers.
To conclude, Dejan Juric showed that ALP + FUL treatment was beneficial in patients with HR+, HER2– advanced BC, especially with a low TMB, but that a comprehensive understanding of the unique mutational profile of each tumor via biomarkers analysis may explain the level of success and thus dictate further care.
Trastuzumab deruxtecan versus treatment of physician’s choice in patients with HER2-low unresectable and/or metastatic breast cancer: Results of DESTINY-Breast04, a randomized, phase 3 study3.
Shanu Modi, Memorial Sloan Kettering Cancer Center, Memorial Hospital, New York, NY, USA
Metastatic breast cancers (mBC) are classified according to the detection of certain receptors in the tumor cells, dictating the type of treatment to offer to the patients. Thus, mBC with an abnormally high quantity of human epidermal growth factor receptor-2 (HER2+) benefit from therapies targeting HER2 protein with monoclonal antibodies, while HER2- mBC receive treatment based on their HR status. However, the dichotomy between HER2+ and HER2- mBC does not suffice to find effective therapies for patients with low level of HER2 (HER2-low) currently treated as HER2-. The limited options and modest benefits of chemotherapy confirm the need for an adapted targeted strategy.
Trastuzumab deruxtecan (T-DXd) is part of a new generation of antibody-drug conjugates that delivers precision-focused chemotherapy directly to the cancer cells. Its activity was shown in tumors across a broad range of HER2 expression and a phase 1 trial showed promising efficacy of T-DXd in patients previously heavily treated with HER2-low mBC. Here, Shanu Modi presented us the DESTINY-Breast04 study, the first randomized phase 3 study of T-DXd for HER2-low mBC and its auspicious results.
Measuring the median progression-free survival (mPFS) in HR-positive mBC patients as primary endpoint, they observed statistically significant and clinically meaningful improvement for patients with HER2-low mBC compared to standard chemotherapy (mPFS 10.1 versus 5.4 months respectively: p<0.0001). Similar benefit was seen in all patients, regardless of their HR status, with T-DXd treatment through PFS and overall survival (OS) compared to standard chemotherapy.
DESTINY-Breast04 establishes HER2-low mBC as a targetable patient population with T-DXd as a new standard of care, with the potential to improve the survival for ~50% of all mBC patients.
We expanded the benefits of HER2 targeted therapy to a new population of breast cancer patients and established T-DXd as the new standard of care for HER2-low mBC.
These ground-breaking results presented at the 2022 ASCO Annual Meeting, and simultaneously published in the New England Journal of Medicine4, were acknowledged with a standing ovation from the audience of specialists. Anticipating these results to be practice changing, the study gives hope for many oncology professionals and patients.
Circulating tumor DNA and late recurrence in high-risk, hormone receptor–positive, HER2-negative breast cancer (CHiRP)5.
Marla Lipsyc-Sharf, Dana-Farber Cancer Institute, Boston, MA
Over half of metastatic recurrences in HR+ BC are late (occurring over 5 years from diagnosis) and thought to arise from minimal residual disease (MRD), a small number of cancer cells left in the body after treatment, hence the benefit of adjuvant therapy. MRD detection via circulating tumor DNA (ctDNA) is associated with high risk of BC recurrence in the early adjuvant setting across tumor subtypes. Little is known, however, about ctDNA for later settings.
Marla Lipsyc-Sharf presented the CHiRP prospective study of late recurrence in patients with high-risk HR+ BC without prior recurrence. 83 patients were followed with whole exome sequencing on primary tumor samples and plasma collection every 6-12 months to be processed with personalized RaDaRTM assay (12-51 variants) to detect ctDNA. Patients did not undergo routine surveillance body imaging or other circulating biomarkers testing. 68.7% of the patients had stage 3 disease and most received chemotherapy (90.4%) and adjuvant endocrine therapy (100%).
8 of 83 (10%) patients had detectable ctDNA at any timepoint during this study. As of last follow-up, 6 of them developed metastatic recurrence at various sites, 6-14 years after primary diagnosis, and one patient with detected ctDNA developed a locoregional recurrence.
All distant recurrences were detectable via ctDNA prior the recurrence with a median lead time of ~1 year.
Despite the low yet steady rate of recurrence in this small cohort with limited follow-up and infrequent plasma sampling (every 6-12 months), this study, published in Journal of Clinical Oncology6, shows that liquid biopsy can provide precious indication on the risk of relapse and thus point towards earlier intervention after MDR detection, improving patients' survival and quality of life.
1 https://meetings.asco.org/abstracts-presentations/210258
2 https://meetings.asco.org/abstracts-presentations/209230
3 https://meetings.asco.org/abstracts-presentations/209021
4 Modi S, Jacot W, Yamashita T, et al. Trastuzumab Deruxtecan in Previously Treated HER2-Low Advanced Breast Cancer [published online ahead of print, 2022 Jun 5]. N Engl J Med. 2022;10.1056/NEJMoa2203690. doi:10.1056/NEJMoa2203690
5 https://meetings.asco.org/abstracts-presentations/209216
6 Lipsyc-Sharf M, de Bruin EC, Santos K, et al. Circulating Tumor DNA and Late Recurrence in High-Risk Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Breast Cancer [published online ahead of print, 2022 Jun 4]. J Clin Oncol. 2022;JCO2200908. doi:10.1200/JCO.22.00908
SOPHiA GENETICS products are for Research Use Only and not for use in diagnostic procedures unless specified otherwise.
SOPHiA DDM™ Dx Hereditary Cancer Solution, SOPHiA DDM™ Dx RNAtarget Oncology Solution and SOPHiA DDM™ Dx Homologous Recombination Deficiency Solution are available as CE-IVD products for In Vitro Diagnostic Use in the European Economic Area (EEA), the United Kingdom and Switzerland. SOPHiA DDM™ Dx Myeloid Solution and SOPHiA DDM™ Dx Solid Tumor Solution are available as CE-IVD products for In Vitro Diagnostic Use in the EEA, the United Kingdom, Switzerland, and Israel. Information about products that may or may not be available in different countries and if applicable, may or may not have received approval or market clearance by a governmental regulatory body for different indications for use. Please contact us to obtain the appropriate product information for your country of residence.
All third-party trademarks listed by SOPHiA GENETICS remain the property of their respective owners. Unless specifically identified as such, SOPHiA GENETICS’ use of third-party trademarks does not indicate any relationship, sponsorship, or endorsement between SOPHiA GENETICS and the owners of these trademarks. Any references by SOPHiA GENETICS to third-party trademarks is to identify the corresponding third-party goods and/or services and shall be considered nominative fair use under the trademark law.