Why We Need Big Infrastructure For Tackling Rare Disease

Hannes Smarason big data rare diseases

Having bigger databases, and having a mechanism that flags genomic variants, is key to optimizing patient care for entire families

Uncovering the genetics of schizophrenia is vital but challenging. As I wrote in my last post, mutations in more than 100 spots in the genome have been linked to the condition. But which ones actually play a role in the disease, and which ones are just there for the ride—innocent bystanders that just happen to occur alongside the real culprits? That’s the crucial question for scientists seeking new treatments for this condition, among them leading researchers and clinicians at our close partner, Boston Children’s Hospital (BCH).

One thing we’ve learned recently is that even a small amount of knowledge about genetic underpinnings of disease can have a big potential benefit for patients. For example, the 16p13.11 region deletion I described in that last post ended up being very important for several patients later, particularly one father and his son, recently described by our colleagues at BCH. This case highlights the importance of expanding the scope and scale of such research, and of updating and alerting patients as more is discovered—not just in schizophrenia, but across rare disease.

In their previous work, the BCH team used chromosomal microarray analysis to determine that a young boy with symptoms of schizophrenia, including psychosis, was missing an entire chunk of DNA—one copy of the chromosomal region 16p13.11, which spans several genes.

Schizophrenia in children is rare, and some researchers believe it could be an extreme variation of the disease, and so might hold important clues for the treatment of this condition in both young and old. A search of our and BCH’s databases showed that several other young patients also showed variations in that region. Just as important, it was confirmed that a parent of one of those patients also carried that deletion, and it seemed likely that another parent (not available for testing, but with reported symptoms of schizophrenia) also likely carried the deletion.

Clearly 16p13.11 seemed to be emerging as a “hotspot” for variations linked to psychosis. But the scientists were only finding this because they could go back and search the databases, and they were working their way backwards from pediatric cases to learn information that might have been medically relevant to the parents as well. All this suggests that having bigger databases, and having a mechanism that flags such variants, is key to optimizing patient care for entire families.

One case uncovered by the BCH scientists, regarding a young man who we will call Jack, brought this into sharp focus. As a teenager, Jack had undergone detailed genetic screening at BCH because of symptoms that included learning disabilities and recurrent seizures. It was determined that he had a 16p13.11 deletion, but at the time of his screening, that mutation hadn’t yet been linked to psychosis. So it became just one more detail in Jack’s medical record.

Separately, a few years later, Jack’s father was diagnosed with ADHD and treated with a high dose of mixed amphetamine salts. Within a few weeks Jack’s father experienced a manic-psychotic episode. He was prescribed an anti-psychotic and eventually recovered. Unfortunately, his son was deeply affected by his father’s breakdown and became withdrawn and depressed. Eventually, Jack also developed psychotic symptoms, which were so serious he was hospitalized.

Jack’s symptoms, thankfully, responded to anti-psychotic medication, but his doctors wondered if there was any connection between the breakdowns suffered by the father and son.

A check of Jack’s medical record revealed the 16p13.11 deletion. And seeing that detail after the link had been made between 16p13.11 and psychosis, his doctors immediately speculated that it might be a cause of Jack’s symptoms. Further, they suspected that mutation could be the “linchpin” causing psychosis in the father and the son. Jack’s father was tested, and he also carries a 16p13.11 deletion.

So here’s the lesson: if Jack’s doctors had known about the link between 16p13.11 and psychosis as soon as it emerged, they might have also suggested testing Jack’s father. If they had, the BCH doctors “believe that the psychosis could have been averted in both father and son.”

In light of this case, the BCH researchers write that they see a keen need for broad, integrated, and sophisticated infrastructure to support genomics-driven precision medicine. They have several recommendations, including that physicians need to receive regularly updated risk information about specific mutations; genetic reports on parents who are “carriers” but seem unaffected should note that problems could arise later, and families that include carriers of variations that increase risk should be monitored and given counseling.

Such activities will be well supported by tools such as WuXi NextCODE’s Genomically Ordered Relational (GOR) database and global platform for diagnosing rare disease and building a global knowledgebase. This can act as one of the key spokes in the “wheel” of genomic diagnostic process. But we also need to build in others, such as means to automatically alert doctors to important knowledge updates, monitor patient records, and connect doctors to specialists who can help refine a diagnosis as new discoveries are made. We and our partners at BCH are committed to helping create these tools.

*  *  *  *  *  *  *  *  *  *  *  *

Headed to ASHG? If you are attending ASHG this month, join us to hear more about how rare disease studies can inform our understand of common diseases at two of our “Genomes for Breakfast” events: “Using Population Genomics to Understand Common and Rare Disease” (Oct. 18), and “Using NGS to Diagnose Rare Disease—Experiences from Three Continents” (Oct. 19).

email

New Breast Cancer Study Underscores the Need for More Sequencing

Gene sequencing for breast cancer. More than the usual suspects at play.

Ever since actress Angelina Jolie’s highly publicized preventive mastectomy ignited discussion about BRCA 1 and BRCA2, almost everyone has heard about these genes and how they can increase risk of breast cancer.  Some people even refer to them as “the breast cancer genes.” But how genes cause this disease is much more complicated than just through the most well known BRCA mutations, as a recent study in JAMA of Ashkenazi Jewish women has demonstrated. http://jamanetwork.com/journals/jamaoncology/fullarticle/2644652

This intriguing study raises a crucial question: How much sequencing is enough when diagnosing breast cancer in the age of targeted therapies? The number of these therapies keeps growing, as does our knowledge of the links between what drugs work for women with particular mutations. But at what point should we say we have uncovered enough mutations to make a proper diagnosis? And in a field in which we know there’s a lot we don’t know, is there such a thing as enough?

The good thing is that sequencing costs are going down. “It used to be that just testing for a single gene cost several thousand dollars,” says Jim Lund, Director of Tumor Product Development at WuXi NextCODE.  “Now a panel of genes costs that and whole exome sequencing is slightly more.” At the same time, the number of mutations that are discovered and studied is increasing – in the genomes of patients and the genomes of their tumors.

The data here has a message about data itself: in principle, we should be generating as much sequencing data as possible. By generating it, storing it for vast numbers of patients and their healthy relatives, creating more comprehensive databases of all disease-linked variants, and then reanalyzing patient and tumor samples as more is learned, we can improve the risk assessment and the speed and accuracy of diagnosis for patients everywhere. Since we can do this, the question isn’t whether we can afford to do more sequencing, but why anyone would argue that we can afford not to.

The researchers who led the recent JAMA study used multiplex genomic sequencing on breast tumor samples from 1007 patients. They tested for a total of 23 known and candidate genes.  It has been well documented that women of Ashkenazi descent have a higher risk of breast and ovarian cancer, and that is at least in part because of three particular BRCA1 and BRCA2 mutations. These are called founder mutations, because they probably originated among some of the earliest members of this ethnic group, and have been propagated because of a strong history of marriage within the same community.

But the researchers working on this study wanted to know if there were mutations in other genes besides BRCA that made it more likely these particular women would develop breast cancer. The patients were from 12 major cancer centers; had a first diagnosis of invasive breast cancer; self-identified as having Ashkenazi Jewish ancestry; and had all participated in the New York Breast Cancer Study (NYBCS).

Surprisingly, only 104 of the patients were carrying one of the infamous founder alleles. Seven patients had non-founder mutations in BRCA1 or BRCA2, and 31 had mutations in other genes linked to increased risk of breast cancer, including CHEK2. The vast majority of these women carried none of the mutations that are “obvious suspects” for breast cancer. “We do not know why those women got breast cancer,” says Shannon T. Bailey, Associate Director of Cancer Genetics at WuXi NextCODE.

It’s important to note that thousands of different cancer-predisposing mutations have been found in BRCA1 and BRCA2 alone. Every population studied to date includes people with such mutations.  The three founder mutations that have been established as being common among Ashkenazis are estimated to account for about 10% of breast cancers in this group. The rest of BRCA1 and BRCA2 mutations are considered extremely rare under any circumstances.

“If you look at the genes on the panel used in this study, it looks as if they are mostly associated with DNA damage and there are no cell cycle regulating genes included,” says Bailey. “But there are all kinds of mutations that cause breast cancer, even in noncoding regulatory zones.” As a result, even the best designed panel may fall short.

That’s why this study is so important. It tells us that even with founder mutations, family history matters but it doesn’t yet always tell you everything you’d like to know. Of the women with the founder BRCA mutations, only about half had a mother or sister with breast or ovarian cancer.  It’s also already well known that just carrying a BRCA1 or BRCA2 mutation is no guarantee the patient will get cancer. For reasons we don’t yet understand, these mutations raise overall risk, but not everyone who carries one will develop the disease. So while BRCA mutations are important, we need lots more information about other genes too.

The authors of this JAMA report suggest that Ashkenazi patients with breast cancers should have “comprehensive sequencing,” including, perhaps, complete sequencing of BRCA1 and BRCA2 and possibly testing for other breast cancer genes as well.

And what about other patients?  WuXi NextCODE’s Lund points out that even the most highly regarded recommendations for breast cancer testing do not cite specific panels. Those recommendations come from the U.S. Government Task Force [https://www.uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/brca-related-cancer-risk-assessment-genetic-counseling-and-genetic-testing] and the NCCN Clinical Practice Guidelines. Women with a family history will likely get more comprehensive testing, but beyond that it is not clear exactly how to proceed in every case.

At WuXi NextCODE we believe that this is clear evidence pointing to the value of doing more sequencing across all ethnic groups – for healthy individuals, patients, and their tumors, and pushing towards sequencing as standard of care. This would expand our knowledge of the genetic risk factors for breast and other cancers; provide vast new cohorts for research; and deliver the most actionable insights to patients, from risk assessment through diagnosis and then ongoing as new discoveries are made.

All of the participants in this JAMA study consented to have their sequence data used to advance research. They are already helping to do that, and this is just one study of thousands that are now underway and that are helping us to expand our data- and knowledgebases with the ultimate aim of delivering even better outcomes for all people and patients everywhere.

Let’s Speed the Genomic Revolution, UK CMO Says

Sally Davies genomics

Whatever path various societies take to tap the power of the genome to improve human health, a recent report from England’s Chief Medical Officer, Dame Sally Davies, calls out key elements for realizing that future sooner rather than later.

England’s Chief Medical Officer wants to build on the success of Genomics England’s 100,000 Genomes Project and take her country swiftly into the age of precision medicine. The goal is to get patients optimal treatment more quickly and with fewer side effects. That means using genomics to more accurately guide prescribing—initially for cancer, infections, and rare diseases—but increasingly for all conditions and overall wellness and prevention.

Dame Sally Davies’ vision is anchored in the work that Genomics England is engaged in today and to which WuXi NextCODE and other leading genomics organizations have contributed. It’s a rallying cry that many voices are joining and underpins our work not only in England, but also similar efforts we are helping to advance in countries near and far, from Ireland to Singapore.

Her call is particularly forceful in three areas that she rightly singles out as critical to realizing the potential of precision medicine to revolutionize healthcare:

  • Industrial scale: Genomics has in many ways been treated and developed as a “cottage industry,” yielding important advances. But the need is massive scale in the era of population health (e.g., whole-genome sequencing, or WGS).
  • Privacy AND data sharing: Dame Sally wants to provide and ensure high standards of privacy protection for genomic data but is adamant that this should not come at the price of stifling the data sharing and large-scale collaboration that will transform medical care and many patients’ lives. She wants to move beyond “genetic exceptionalism,” which holds that genomic data is fundamentally different or more valuable than other data. Like other sensitive data, we can protect genomic data well and use it for public benefit.
  • Public engagement: She calls for a new “social contract” in which we, as individuals and members of society, recognize that all of us will benefit if we allow data about our genomes to be studied. That holds whether we are talking within our own countries or globally.

In England, as elsewhere, these shifts require the input of political leaders, regulators, and a range of healthcare professionals, including researchers as well as care providers. Crucially, such a transformation also requires a level of commitment on the part of patients throughout the National Health Service (NHS) and citizenry in general. If England takes this bold step forward, it could have tremendous effects. But “NHS must act fast to keep its place at the forefront of global science,” said Davies. “This technology has the potential to change medicine forever.”

To date, more than 30,000 people have had their genomes sequenced as part of the 100,000 Genomes Project. But there are 55 million people in the UK, and Dame Sally would like to see genomic testing become as normal as blood tests and biopsies for cancer patients: She wants to “democratize” genomic medicine, making it available to every patient that needs it.

We share and are, indeed, taking part in helping to realize much of Dame Sally’s vision as we work to accelerate Genomics England’s work and engage with our partners globally. As we know, different societies have different models of healthcare and different approaches to research and care delivery. But the ability for people anywhere to tap into the power of the genome to improve their health is at the very core of our own mission as an organization, and we applaud Dame Sally for calling out some of the key elements for realizing that future sooner rather than later.

Whatever path different societies choose to follow toward precision medicine, her recent report provides one enlightening view of a starting point for making the leap.

A Perfect Pairing? AI and Precision Medicine

AI-and-precision-medicineLet’s start with one of the fastest-growing fields in science today: artificial intelligence (AI). Now, let’s apply another technology that has profound potential for improving patient care:  precision medicine. Some of us think the integration of these two arenas could be a “sweet spot” that leads to some of the decade’s biggest advances in healthcare.

As someone who has worked in genomics for two decades, I am a believer in the combined power of AI and precision medicine. And in my current work, I have the pleasure of pioneering both technologies.

Cancer has been one of the early beachheads for precision medicine, Now, AI is also following that path, with the aim of advancing individualized treatment.

For example, just today, WuXi NextCODE presented preliminary data from analyses using our novel AI technology at the American Association of Cancer Research annual meeting in Washington D.C. We tested the accuracy of our new deepCODE deep learning tools to diagnose subtypes of tumors. Our results suggest these tools do a better job than traditional approaches for classifying tumors and helping determine which patients will respond to which drugs.  Our new AI technology can incorporate all types of omic data, and can also help with drug discovery and finding the best uses for drugs.

How can AI technologies achieve better results in identifying precision treatments in cancer and other diseases?  In the case of our new deepCODE tools, it is in part thanks to a novel, multinomial statistical-learning method and deep learning classification strategy. This approach is designed to support dramatic improvements in drug discovery and development, as well as medical care. But we need to prove the technology’s potential by testing it on real problems in genomic medicine. So, that’s what we are doing.

The initial results are promising. Our deepCODE tools were validated on six patient-derived tumor xenografts from mouse models, and then tested against approximately 9,000 human tumors from a collection of 27 types in The National Cancer Institute’s Cancer Genome Atlas (TCGA) collection. (https://cancergenome.nih.gov/)  We achieved 95% accuracy overall in this test. In analyses of human breast and lung cancer subtypes, deepCODE was accurate in 96% and 99% of cases, respectively. That study included DNA- and RNA-seq data.

These findings are very encouraging.  Breast and lung cancer are both very common malignancies that are increasingly being “divided” into subtypes that have significantly different outcomes and need different treatment regimens. These preliminary data are by no means definitive, but they suggest that AI could bring new certainty to cancer diagnosis.

But why is it even so important to get a fast, accurate molecular diagnosis of a tumor?

Well, here’s the challenge: Today patients who have suspected cancers are typically biopsied.  A snip of the tumor is examined under a microscope and then may be tested for common biological receptors. It can take a while for that to occur. Next, the patient undergoes treatment, and whatever drugs they receive could actually change the tumor’s biology: After that, the drugs initially prescribed might not be the best option anymore.

So how can we know when to switch treatments, and what to switch to?

In the ideal world, anyone diagnosed with cancer would be followed up with an extensive molecular biopsy. In other words, once the initial diagnosis is made, the patient would undergo follow-up tests that involve relatively painless blood draws. From these blood samples (liquid biopsies), the tumor’s DNA would be read, and that would determine how to best monitor and prescribe for that particular patient going forward.

It is an exciting time to be working on integrating AI technology with the primary tools for improving precision medicine in cancer and other diseases.  We’re just at the start of this journey, and we’ll likely find many other ways that AI technology can impact patient healthcare.

Join us here as we follow this intriguing program’s progress.