Author: Mirjana Phillips

Cancer care varies widely across different communities, but cancer centers struggle to reduce this disparity. One major obstacle is the lack of access to cancer-related data, such as stage, treatment, biomarker testing, and other indicators, from the unstructured EMR documents. Cancer registries may provide some data, but they are often outdated or incomplete, especially for biomarkers, later lines of therapy, and advanced treatments like clinical trials or CAR-T therapies.

The evolution of cancer-specific Natural Language Processing (AI/NLP) has greatly enhanced electronic medical record analysis, enabling the discovery of care patterns. However, not all AI solutions are created equal. OncoLens’ approach seamlessly integrates AI tools with continually updated specialized oncology knowledge models, enabling the identification of oncology-specific keywords, contexts, and patterns. This integration yields valuable, actionable insights, facilitating timely interventions and significantly impacting patient care.

Some examples include:

• Finding cancer staging data — Staging information can now be extracted from unstructured fields, clinical notes and pdfs. AI systems can and must be trained to look for AJCC staging or other staging methods like BCLC (Barcelona Clinic Liver Cancer) or FIGO, along with the time of staging. This information is foundational in oncology to identify quality improvement opportunities and care patterns.
• Clinical trial patient identification — Specific trials require extracting key clinical information. For example, the ADAURA2 trial requires identification of patients who have had a complete resection of the tumor and not just “biopsy” or wedge resections, in addition to other criteria like excluding Squamous Cell histologies and early identification of patients with stage IA2 or IA3 disease. This cohort needs to be tested for EGFR mutations early to see if they are eligible for this trial. Today this extremely manual process can be streamlined with AI/NLP performing the prescreening of patients, savings tens of hours per month in identifying patient cohorts for uncommon disease type trials.
• Identification of patients with particular histologies for genetics counselling — Identification of patients with specific cancer types, like pancreatic cancer and prostate cancer, who have historically not had access to genetic counseling can be identified with the help of real time trackable reports. For instance, patients with metastatic prostate cancer or high or very high-risk prostate cancer and patients with strong family history of other cancers need to be screened for prostate cancer susceptibility genes. Likewise, all patients diagnosed with exocrine pancreatic cancer need to be tested for pancreatic cancer susceptibility genes including ATM, BRCA, CDKN2A and Lynch Syndrome genes.
• Patient identification for Internal Pharmacy Referral — Early identification of patients who should be referred to the Cancer Center pharmacy program for faster, higher quality access to care, including counseling, as in subsets of patients with lung cancer who may need Ipilimumab, pembrolizumab, cancers requiring CAR-T therapies or cancer patients receiving chemotherapy that requires growth factor support with pegfilgrastim.

A visual of different types of search criteria is given below.

Cancer centers call out these top three operational reasons behind disappointing clinical trial accrual numbers:

1. Increasingly complex inclusion/exclusion criteria, very often with targeted biomarker testing requirements that are not standard of care. Providers who are not Principal Investigators are often unaware of these requirements resulting in insufficient testing to confirm eligibility.
2. Poor timing in identifying eligible patients. Complex selection criteria requires more manual hours to be spent by research teams combing through medical records. This results in delayed reporting to providers, after they’ve discussed a different course of treatment with the patient, making it difficult to change course to a trial.
3. Silos of care even within oncology service lines. Increased provider specialization results in silos where surgeons are often not aware of clinical trial options or lung cancer specialists are not aware of biomarker driven, pan cancer trial.

As a trial sponsor, you can see the impact of these challenges on your accrual numbers. OncoLens’ AI driven Patient ID and Collaboration tools enable cancer centers to overcome these challenges.

Cancer programs are required to comply with quality metrics established by accrediting bodies such as the Commission on Cancer (CoC), National Accreditation Program of Breast Cancer (NAPBC), National Accreditation Program of Rectal Cancer (NAPRC), and ASCO QOPI, among others. The metrics are being constantly updated as cancer related data increases. Some old standards are retired, and new metrics are added. These metrics add complexity to workflows and extend the time required for data collection and organization.

As an illustration, on July 28, 2023, the Commission on Cancer’s National Cancer Data Base (NCDB) RCRS introduced updates to three Quality Measures:

• Breast BCSRT: Radiation therapy, when administered, is initiated less than or equal to 60 days of definitive surgery for patients receiving breast conserving surgery for Stage I-III breast cancer who do not undergo adjuvant chemo- or immuno-therapy.
• Gastric G16RLN: At least 16 regional lymph nodes are removed and pathologically examined for patients with surgically resected gastric adenocarcinoma undergoing curative intent therapy.
• Lung LCT: Systemic chemotherapy, immunotherapy or targeted therapy is administered or recommended within 3 months preoperatively or 3 months postoperatively for surgically resected cases with pathologic T2 greater than 4cm or T greater than or equal to 3, or N greater than or equal to 1 Non-Small Cell Lung Cancer (NSCLC).

Overcoming Implementation Challenges with AI

Implementing these updates necessitates investments in training data extractors, creating new databases, and devising actionable strategies to address any identified shortcomings promptly. All cases of the above cancer types must be looked at again, appropriately staged, treatment options collected, dates of discussion reviewed, treatment intent studied and placed in a database for quality metrics calculation.

Leveraging advancements in Artificial Intelligence (AI) can significantly simplify these processes. AI can assist in extracting data from Electronic Medical Records and other sources, making the extraction of metrics more efficient. Following quality assurance checks, the extracted data can be presented to the Cancer Committee for approval and appropriate follow-up steps. Educational efforts can also be seamlessly integrated into the Cancer Conference workflow, enhancing care for patients and the cancer care team.

Subsequent articles will delve into the details of these changes. In the interim, we encourage cancer teams to explore the potential of automating cancer quality metric extraction through AI. Connect with us at info@oncolens.com to learn more about this transformative approach.

For inquiries about the new quality measures from the CoC or program-specific questions, reach out to the CAnswer Forum. Contact ncdb@facs.org for program-specific queries.

Clinical trials play a pivotal role in bringing new, more effective cancer treatments to market, yet less than 5% of cancer patients enroll in clinical trials. This makes it difficult for pharmaceutical companies to meet the recruiting requirements to complete their clinical research and keeps potentially life-saving therapies out of reach of patients who could benefit from them. 

So how can we increase clinical trial participation and introduce new drugs to all of the patients who need them? Community-based clinical trials could be the answer.

Pharmaceutical companies have relied on large cancer centers and academic medical centers to host their clinical trials for decades. These institutions are typically in densely populated areas and see a large number of patients, making them attractive to companies conducting research studies. 70% of patients, however, live over two hours from the closest academic medical center and a recent Labcorp study¹ indicates that cancer patients are exponentially less likely to participate in a clinical trial if they have to drive over 25 miles. Community-based trials address this problem by holding trials at smaller, local sites instead of only at large academic medical centers.

By casting a wider net, pharmaceutical companies also engage a more diverse patient cohort in their clinical trials – which helps meet regulatory guidelines and informs drug manufacturers of the efficacy of their drugs on broader patient populations. For local communities, clinical trials bring economic benefits to the area as well as more options for quality healthcare, including new and cutting-edge treatments.

Pharmaceutical companies have realized the benefits of shifting to a more decentralized approach to clinical trials and the community-based trend is becoming the norm. 77% of all current clinical trials occur in a community setting.  Not including the 20 largest hospitals and academic medical centers, 85% of trials are now conducted in a community setting. This trend impacts sponsors, local communities, and patients.

For sponsors, community-based trials can help with recruitment and diversity, two areas that have traditionally been a challenge for pharmaceutical companies. Since roughly 20% of new drugs have different effects on people of different races, sponsors must recruit diverse clinical trial participants to ensure that their treatments are safe and effective.

But recruiting patients into trials is no easy task. According to the NIH, over 80% of clinical trials experience delays because of recruiting challenges² and 85% of clinical trials fail to hit their recruitment goals.

There are many reasons that factor into patients declining to participate in clinical trials. They may:

• Not be able to take off work
• Not have a reliable means of transportation
• Not be able to afford childcare during their treatment
• Not feel healthy enough to travel long distances

Community-based clinical trials can eliminate many of these obstacles. Participants can visit a local community hospital, doctor’s office, or lab instead of driving hours to visit the main study site. This limits the time a patient has to take off from work and makes transportation much more convenient. To increase recruitment and accrue an appropriately representative patient cohort for their clinical trials, sponsors need to reach the 70% of patients who don’t live near an academic medical center.

Clinical trials offer new treatment options that many underserved areas and underrepresented patient populations have previously not had access to. By participating in trials, health systems and physicians bring these options to their local communities and offer their patients access to higher quality healthcare, including cutting-edge treatments and therapies.

By extended clinical trials to the community setting, sponsors gain access to a new pool of participants in need of treatment. Local health systems benefit from the funding and medical expertise flowing into their communities.  And most importantly, patients gain more and better options for their healthcare. Options which could save their lives.

The trend towards community-based clinical trials is a positive movement for pharmaceutical companies, health systems, and patients alike. These smaller sites, however, often face staffing and infrastructure challenges that are less burdensome for larger academic medical centers or cancer centers. With an average of 66 trials at community sites, it can be challenging for physicians to keep track of which trials are available at their sites. And with over 30,000 ongoing clinical trials in oncology³, it’s impossible to be aware of all options for their patients.

Well, almost impossible.

OncoLens works with over 200 cancer centers in the community setting and partners with clinical trial sponsors to ensure that care teams are always informed of the latest research studies available. Through the OncoLens platform, healthcare providers are alerted to potential trial matches at the point of treatment decision making.

By providing the most relevant information to physicians at the most appropriate time, OncoLens helps providers consider all possible options and make the most informed decisions for their patients. This helps tremendously in the identification and recruitment of patients into clinical trials. Given the time-sensitivity of oncology treatment decisions, the ability of healthcare providers to quickly match patients to clinical trials can potentially be the difference between life and death.

With more and more trials being conducted in the community setting, OncoLens aims to take the guess work out of the equation for community providers so that they can provide the highest possible quality of care.  Oftentimes, that means informing them of the latest clinical trials at the right time in their patient’s journey.  And the right time is as early as possible.

References

¹ https://drugdevelopment.labcorp.com/content/dam/covance/assetLibrary/whitepapers/Understanding-New-Paradigm-WPDCT001.pdf

² https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7342339/

³ https://clinicaltrials.gov/ct2/results?cond=oncology&Search=Apply&recrs=b&recrs=a&recrs=f&recrs=d&age_v=&gndr=&type=&rslt=

Despite the 2016 launch—and 2022 reignition—of the National Cancer Institute’s Cancer Moonshot^SM and the 2015 Precision Medicine Initiative, Molecular Tumor Boards to direct personalized cancer care still are few and far between. A few reasons drive that scarcity including the limited availability of molecular pathology specialists, time of preparation and research required for identification of appropriate therapies and the effort to bring a multi-disciplinary team together. OncoLens’ tumor board solutions bring efficiency and ease of access to providers and ultimately patients. Biomarker data is automatically integrated along with the patient’s clinical data and discussions can be easily extended to include the right experts and cases sourced across health system or practices.

The New Molecular Recommendation Builder from OncoLens provides the infrastructure necessary to enable institutions to further standardize their Molecular Tumor Boards. It enables them to record and search for prior recommendations made by biomarker, creating a database of targeted therapies and clinical trials. In essence, it enables the development of the center’s own pathways. The module includes a streamlined workflow to search across internal prior recommendations and external best practices—all the way through to development of a final recommendation letter.

In an Annals of Oncology article on “Molecular Tumor Boards: current practice and future needs,”¹ the authors write that “MTBs are critical to close the growing gap between clinical practice and technological potential in cancer care,” and they note the limitations of standard pathology and sequencing reports that report only mutations with well-known treatment consequences while overlooking those for which alternative treatments might be available. In light of their conclusion that “Since the opportunities for (and, consequently, complexity of) genomic tumor testing and genetically guided treatment are continuously expanding, patients rightfully expect that treatment opportunities are recognized and discussed.” MTBs can help make that complex genomic information actionable, supporting patient-centered treatment decisions. The Molecular Recommendation Builder can make it easier for MTB members to visualize the biomarker information and make evidence-based recommendations based on prior recommendations or best practices.

Accessible directly from the case, Molecular Recommendation Builder matches the patient’s genomic results with potential therapies for their specific clinical picture, based on best-practice guidelines and a proprietary institutional database of guidelines, eliminating the need to research and identify similar case histories in disparate locations. MTB members can add to or edit matching recommendations based on the current patient, linking to clinical trials and references. This creates a new set of recommendations for the patient while continuing to expand the institutional database, enhancing the recommendation suggestions for future patients.

As genomic sequencing becomes the standard of care, molecular tumor boards will play a key role in enhancing our understanding of molecular characteristics and knowledge of the therapeutic approaches with the best possible outcomes. With the Molecular Recommendation Builder, OncoLens makes it possible for MTBs to demonstrate their expertise, build their knowledge base, and reach more patients.

———-

¹ van der Velden DL, van Herpen CML, van Laarhoven HWM, Smit EF, Groen HJM, Willems SM, Nederlof PM, Langenberg MHG, Cuppen E, Sleijfer S, Steeghs N, Voest EE. Molecular Tumor Boards: current practice and future needs. Ann Oncol. 2017 Dec 1;28(12):3070-3075. doi: 10.1093/annonc/mdx528. PMID: 29045504.