Improve Your Cancer Treatment with Personalized Data
Summary: Getting to your best cancer treatment depends on accessing lots of data about you and providing it to experts and expert systems to get their recommendations. Unfortunately, getting access to and managing the large volumes of your health data has been slow, awkward, and complex. Fortunately, there are new apps for accessing and managing your health data, enabling you to proactively participate in what used to be a closed system.
The More Health Data You Have, the Better
Data is the feedstock and foundation for the explosion of unique treatment solutions for everyone (“precision medicine” and “personalized medicine”). If you have a cancer diagnosis, you may be lucky and have a “tried and true” treatment that lines up with your cancer, as it did for me. But for many people with a cancer diagnosis, there are many treatment options, some of which are experimental, and you need to figure out which are the 3 to 5 best cancer treatment options for you. To identify your best options, you need to get as much data about you as possible, yet the data are often spread out, or in difficult-to-decipher doctor’s notes or reports in medical record systems. Once gathered, your data needs to be easily available to experts and expert systems for the matching process, and to medical researchers to advance therapies.
It’s Slow, Awkward, and Complex to Manage Your Health Data
The current systems for getting access to and managing the large volumes of your health data are slow, awkward, and complex.
· Slow: It takes a long time to pull together and analyze all your relevant health data while you and your loved ones are feeling scared, depressed, anxious, pre-occupied, and hyper-aware that the clock is ticking in response to a cancer diagnosis.
· Awkward: It’s harder than you might expect to overcome the barriers to accessing, assembling, and linking all of your potentially useful health data.
· Complex: It’s overwhelmingly complicated to sort through all the potentially useful data to extract the key data nuggets that are essential to driving critical insights.
Consider these stories of data management challenges suffered by people with cancer diagnoses:
· Slow — It’s hard to get data from your medical tests: In July of 2018 I was diagnosed with lymphoma. My tumor was biopsied twice at a leading cancer center. I asked to have my tumor sequenced many times because I had heard it could be very useful in helping me find my best treatment options, but it didn’t seem to be important and kept being put in the next processing batch. Finally, 10 months later in May 2019, I was informed, after asking repeatedly, that the sequencing had been completed and my oncologist would review the results with me at my next appointment in August.
· Awkward — It’s hard to get medical data from one hospital to another: Former Vice President Joe Biden has railed against the infuriating challenge of getting his son Beau’s medical records from one hospital to another (Walter Reed to MD Anderson) when he was battling a serious brain cancer (Stage 4 glioblastoma). “When my boy was dying, I asked whether or not … we could have access to his cancer genome sequencing in a great hospital. They said, ‘Well, that’ll be hard.’ … Why is it that hospitals are reluctant to allow that data being transferred? Because you might go somewhere else… I’d like to be able to say that there are widely distributed and applied techniques and mechanisms to share data among researchers, clinicians, and patients, but I can’t. I’d like to be able to say that cancer diagnosis in St. Louis is decipherable by another doctor in Memphis, in the case of a patient who moves or seeks a second opinion, but I can’t say that either.”
· Awkward — It’s hard to share data to coordinate care across multiple personal health issues: Shari Mills (I’ve disguised her name) was diagnosed with breast cancer and had breast surgery, which went well. Then she saw her medical oncologist for her next steps, but the oncologist hadn’t looked at Shari’s medical history (which included chronic pain and fatigue) and medications. When the oncologist started talking about a hormone therapy (an estrogen blocker) that causes bone pain, Shari had to tell her that if she took the recommended therapy, she would need to come off one of her essential medications.
· Awkward — It’s hard to transfer large health data files: In March 2017, Steve Aldrich was diagnosed with a cancer (end stage adenocarcinoma in the esophagus) for which there is no known treatment and 90% are dead in the first year. But Steve had done research on the implications of the rapid advance of gene synthesis and gene sequencing technologies in revolutionizing cancer therapies. Craig Ventner of Human Longevity, Inc., (HLI), offered to do a full set of analyses of Steve’s cancer tissue and his biomes. And a German doctor offered to develop a personalized therapeutic vaccine for Steve if the doctor could get Steve’s genetic data, which was big, in many file formats, and scattered around HLI. Steve had to ship hard drives with his health data to Germany.
· Complex — It’s hard to assemble data for a second opinion: Ed Clark (I’ve changed his name) has been helping his wife consider her treatment options since she was diagnosed with ovarian cancer several years ago. She has been through surgery and multiple rounds of chemotherapy. As her options become fewer, they have gotten second and third opinions, searching for clinical trials and hope, including one at Memorial Sloan Kettering, one of the leading academic cancer research centers. As preparation for their visit, Ed spent several days combing through thousands of pages of his wife’s medical records at Kaiser and UCSF to prepare the forms Memorial Sloan Kettering needed for the visit.
Not having access to all your data can result in bad outcomes, such as:
· You can miss personalization opportunities — get treated with a drug that you could know in advance won’t work for you or take a drug that isn’t the best fit for you.
· You can miss out on access to new, less invasive life-saving options, e.g., clinical trials, targeted therapies, off-label drugs. Instead, you will start with the standard of care: tried and true, but usually with harmful side effects.
· You can miss integration (oncology, nutrition, health history, …).
Why Is Managing Your Health Data Increasing in Importance Now?
· Gene sequencing and synthesis technology à genomic data (+other data): There is an explosion of possible health data sources, e.g., scanning of the genes in cells, medical devices for monitoring, and an explosion of cancer treatment options, especially engineered molecular therapies, such as immunotherapies (e.g., CAR-T and TIL). The speed of change in diagnostics, therapies, and the standard of care is accelerating.
· Growing data analysis capabilities (AI and ML, Cloud): There is an explosion of personal health data and therapy options, which has created an increasingly complex matching challenge that can be addressed by new software capabilities (artificial intelligence, machine learning, and cloud computing). The amount of information and complexity is more than healthcare professionals can manage without help from software.
· Entrance of Startups (and Funding), Especially for Consumer Apps: Startups and digital giants (e.g., Alphabet, Apple, Amazon) are developing services, especially consumer apps, that support trusted healthcare teams and individual consumers.
You Should Use Online Services for Managing Your Health Data
Following a cancer diagnosis, it’s hard enough to develop and maintain your care plan, follow recommended therapies, and prepare for physician visits. But living with a cancer diagnosis also means ups and downs as therapies work for a while and knock down the cancer, but the cancer often mutates and recurs. Even people who are in remission need to be prepared for a possible recurrence of their cancer by having a prioritized list of therapies that they will initiate when needed. Preparing your prioritized list of 3 to 5 best treatment options depends on a matching process between an increasingly wide array of treatments that can attack your tumors (surgery, chemotherapy, radiation) and strengthen your defenses (immunotherapy, CAR-T, TIL) with the explosion of possible data sources. The matching process is powered by experts and emerging expert analytical systems. The data foundation to feed into the matching process is built by capturing and sharing as much data as possible from exploding data sources, including your medical history (spread across various medical record systems), gene sequencing (of your tumor, healthy tissues, messenger RNA, and microbiome), and monitoring sensors (e.g., an Apple Watch).
You need help, and fortunately, there are emerging services for managing your health data. To help you in navigating among these services, I’ve organized them into two major considerations (please see the map “Manage My Health Data”):
· Your health data scope: Are you trying to manage your cancer-related data, your DNA data, your medical records data, or all your health data (including sensors and raw genomic data)?
· Your motivation: Are you trying to manage your data to get your personalized list of 3 to 5 best treatment options, or is your goal to provide your data to medical research, or a combination of both?
I’ve further organized your possible online health data management services into three major groups: (1) patient portals for accessing your hospital records, (2) consumer apps for storing and sharing your health data, and (3) services for sharing your health data for research.
1. Access Your Health Data in Your Provider’s Electronic Medical Record System through a Patient Portal
Hospitals and physician practices use electronic medical record systems to track patient visits and get reimbursed for their services. The enterprise software system providers, such as Allscripts, Athena Health, Cerner, Epic, and LinkDoc, offer a portal for the health systems so that you can see your data, such as results of tests, and communicate with your providers. If you are getting your care from one trusted physician or healthcare institution, and you don’t have detailed genomic data or data from wearable devices to add, then this solution may be all that you need. However, the enterprise software system providers haven’t been very good at sharing medical records data between their systems, or between hospital systems, despite demands from physicians and patients. Despite industry efforts to enable “interoperability”, or sharing of data between systems, (a healthcare standards developing organization, Health Level Seven International, has developed standards for data exchange, Fast Healthcare Interoperability Resources, or FHIR, pronounced “fire”), adoption has been slow. And since these systems are primarily used for insurance reimbursement, they are not set up for guiding you to your best cancer treatment options. Lacking your detailed genomic data, mobile telemetry, health behavior data, and data from your medical devices, today’s electronic medical records are inadequate to realize the full potential of personalizing your cancer treatment.
Flatiron Health provides a cancer-focused electronic medical record system (“OncoEMR”) to providers focused on cancer treatment. Their strength is in being tailored for cancer treatment and their efforts to clean and code health data. Flatiron’s mission is to help aggregate health data to move forward cancer treatment research. Since they serve cancer practices, if your trusted cancer provider (oncologist) is using this system, and you don’t have additional health data you want to manage, this solution may be all that you need.
2. Use a Consumer App to Store and Share Your Health Data
· For Managing Your Medical Records: Consumer apps are available that can access electronic medical records from all of your providers, pull it together in one place, and share it with everyone involved in your care, through a mobile phone or web interface, such as Andaman7, Backpack Health, Picnic Health, and Sync.MD. Andaman7 is free for individual patients and providers. All of your health data is stored on your phone (not in the cloud). If your health data is spread across several health institutions, and you don’t have sensor data or raw genomic data, this solution may be useful for you. A note of caution comes from the experiences of Google (Google Health) and Microsoft (HealthVault), which offered similar apps to consumers for managing their medical records. Google Health was shut down in 2012 and Microsoft will be shutting down HealthVault in November. They suffered low adoption in part due to their focus on traditional health records over new sources of data. Electronic medical records are an extremely small part of the overall picture of a person’s health today.
· For Managing Your Medical Records and Sharing with Researchers: Consumer apps such as Ciitizen and Medfusion, not only allow you to collect and manage your medical records from across different health systems, but also allow you to share your data with researchers. If you want to collect then share your medical records data with not only family, friends, and providers, but also researchers, these apps might be useful for you.
· For Managing All Your Health Data: Consumer apps such as b.well, doc.ai, and LifeOmic allow you to manage not only your medical records, which record your visits to providers and test results, but also data from health devices and sensors, such as an Apple Watch or FitBit, or your raw genomic data (not just a summary report). These apps can also help you discover personal health insights with an associated A.I.-powered treatment recommendation engine. b.well serves hospitals, so you will have to access them through your provider. If you want to manage these additional data types, then these apps might be useful for you.
· For Managing and Sharing All Your Health Data: Consumer apps such as the Apple Health Kit, Seqster, and HealthChampion allow you to manage and share all your health data, including importing medical data from health records and adding genetic or fitness data from other sources. If you want to manage and share all health data types, including raw DNA data (not just reports) and health monitoring device data, these apps might be useful for you.
· For Managing Your DNA Data: Companies such as 23andMe and Nebula Genomics analyze your DNA (“sequencing” your normal cells), gather health history information about you by surveying you, and provide personalized reports on variants, risk scores, and relevant research. 23andMe examines a dozen or so genes known to drive a higher risk of certain diseases for around $200. Nebula analyzes a much wider set of genes, and with their novel business model, you will pay nothing or very little, own your data, and may even be able to make money off it. If you want low cost insights into your predisposition for some diseases, these services might be useful for you.
· For Managing Your Cancer Data: Companies such as Belong and CancerAid enable storing and sharing of medical files, plus finding a social network for cancer patients, caregivers, and healthcare professionals. If you want to share your cancer data and combine it with social networking, this may be an option for you.
3. Share Your Health Data for Research
· For Sharing All Your Health Data: Organizations such as All of Us offer services to share your health data with researchers. The All of Us Research Program has set a goal to gather data from one million or more people living in the United States to accelerate research and improve health. By considering individual differences in lifestyle, environment, and biology, researchers hope to uncover better treatment paths through precision medicine.
· For Sharing Your Medical Data: Organizations such as RD.MD in rare diseases allow you to share your data for aggregation by foundations and patient advocacy groups to search for a cure. If you have a rare disease, you should find ways to contribute your medical data to find cures for it.
· For Sharing Your DNA Data: Research programs such as Genomes2People and companies such as Encrypgen and LunaDNA can connect you (and your DNA data) with researchers. The Genomes2People research program at Brigham and Women’s Hospital, the Broad Institute, and Harvard Medical School conducts research in the impact of using genomic information. Encrypgen has a novel business model: you can get paid for sharing your DNA data. LunaDNA also has an innovative business model: it is owned by its community of data contributors. When you contribute and consent to share your data with medical researchers through the LunaDNA platform, you become a co-owner of the organization and share in the value created from medical breakthroughs. If you want to share a wide range of your health data to advance research, you should consider joining these services.
· For Sharing Your Cancer Data: Through research programs such as Count Me In, you can share your medical information, tumor samples, and experiences with researchers to speed the discovery and development of new cancer treatments. You provide copies of your medical records, send saliva and blood samples, and allow access to your stored tumor tissue. Count Me In currently has projects for metastatic breast cancer, metastatic prostate cancer, angiosarcoma, esophageal and stomach cancer, and osteosarcoma.
Get Control of Your Health Data to Help You AND Drive the Healthcare Revolution
Spending your scarce time and money assembling your health data is a challenge. You may not see the need, you may have competing demands, and you may not feel qualified to get directly and personally engaged in your treatment. There is strong evidence from the market that people don’t want to engage directly in managing their health data, such as the retreat by Google and Microsoft from offering consumers a health data repository, and the struggles of startups that have offered direct-to-consumer services, such as Helix (pivoted to B2B2C), Driver (failed), Arivale (failed), and Patients with Power (looking for a buyer).
On the other hand, you may be motivated to actively engage in managing your health data if you have a late stage cancer diagnosis and are searching for the hope of a novel cure, or if you have a rare disease and want to form a community and pool your data with others with the same disease to find a cure, or if you want to contribute or sell your health data to medical research to help others.
Our healthcare system needs a revolution, and people like you, empowered with control over your health data, are the mostly likely drivers. By using the new health data management services described above to manage the increasing complexity you are facing from exploding diagnostic data and treatment options, you can serve not only to improve your treatment, but also accelerate improvement in healthcare. Consider for example the speed with which consumers took up AirBnB, Amazon, and Uber, and disrupted incumbents (hotels, retailers, taxi companies) who were slow to change. Your adoption of these data services will be a spur to hospitals, physicians, and payers, even though it may disrupt their “tried-and-true” processes, overwhelm them with data, face resistance, raise some costs (while saving others), and challenge their traditional patient relationships. By bringing your data together and working with your trusted healthcare expert team in a consultative relationship, you can accelerate the potential for personalizing care for everyone.
Here are some examples of what you can do to improve your treatment with personalized data if you or a loved one has cancer:
· Sign up for and use the patient portal to access your health data on your doctor’s electronic medical record system: My test results are on the electronic medical record system of my academic research cancer center, which I access through their patient portal.
· Enroll in a DNA testing service and respond to their surveys: I was an early adopter of 23andMe. I respond to their surveys. I have yet to get any useful information from them that would influence my behavior, but I feel I’m contributing to better medical insights.
· Search out additional diagnostic tests to build your health profile: As I prepare for the high likelihood of an eventual relapse of my lymphoma, I intend to get more data about my tumor tissue, healthy tissue, RNA transcriptome, and microbiome, put it together in a personal repository, and get expert advice on my 3 to 5 best treatment options. I’ve signed up for a clinical trial with a new blood biopsy company to get personalized biomarkers to track my cancer’s progress. I will release my data to several service providers that specialize in treatment recommendations to get a variety of personalized treatment options.
· Find an app to get all your medical data in one place for analysis: Eric Topol, the doctor and well-known author of several books on disrupting healthcare through the use of data for personalization, used an app (Seqster) to bring together his medical data, sensor (Fitbit data), and test data.
· Sign up for a service to share your data: After my cancer diagnosis, I volunteered to join the All of Us research program. I consented for them to have access to my medical records and went to Brigham and Women’s to give a urine sample and spit sample.
After all this serious analysis, doesn’t that sound like fun?
Disclaimer: In this article I have shared how I am thinking about my treatment, pushing the cutting edge of costly treatment steps that few people will be able to access and navigate. There is no guarantee that the diagnostic data I am going to get — and that I advocate that you get — will make a difference in treatment. And my approach is not intended to take the place of the recommendations of your trusted professional healthcare team; rather I offer it as some things to consider in your deliberations with them.
