If Only There Were a Steve Jobs of Medicine

Since the days of my dad’s company, MultiQuest, I’ve loved creating software. With much excitement, I saw the minor contributions I made to MultiQuest’s software reach an international audience of users. They then used our tool to develop their own complex software more easily and quickly. Since then, I’ve taken a crash course in the biomedical arena and got turned on to medical devices.

MultiQuest's logo - an homage to the early 90s.

I really want to go back to making products, but this time, in the medical device industry. I’ve been a fan of Steve Jobs for a long time. In the past couple of years, I realized that I really want to be like Steve, but for medicine. Steve Jobs’ “simplifying antidote” was really powerful for making computing easier, but in all its complexity, medicine is probably resistant to this antidote.

After Steve’s passing, I read a really well-written post by Forbes’ Matthew Herper on why Steve’s magic doesn’t work in medicine:

Hospitals could use someone to stitch all the gadgets together, and make it all perfect, but there simply may be too much going on for this to happen. [...] Innovation [in medicine] has been technical, and hard to understand, and only physicians and surgeons can grasp it. And we don’t reward it as much as a society: unlike in tech, little of the money in medicine goes to the actual innovator. There are 50% more billionaires from tech than from health care, and they are far richer.

I think there are some great efforts to make the healthcare system simpler, such as with electronic medical records. When it comes to embracing simplifying technologies, though, new devices face issues with reimbursement (discussed a bit more in a previous post, and on Medical Devices Today). Steve Jobs was lucky that he could make products without needing to seek approval from the numerous stakeholders present in medicine (e.g., insurance agencies, doctors, regulatory agencies, government, patients, etc.). Understandably, regulations are stricter on medical devices, but this gave Steve and Co. a lot more freedom to innovate.

Is the difficulty with simplifying medical technologies then more a problem with the healthcare system and policies in this country? Xconomy covered a recent PwC report saying that the US’s medical device leadership is in jeopardy. According to the study’s author, the decline is partly due to these regulatory and policy issues:

In the old world, the FDA was reasonable to deal with, and as long as you followed the rules, you could get an approval [...] Then if you brought it to market, insurance companies would pay for it based on a reasonable assessment of the value you would bring, and the cost it took to bring it to the market. That whole calculus has been thrown out the door.

PwC's Medical Innovation Scorecard. Click the image to see the original report.

The barrier to entry in the US device market is making innovation expensive, and it really seems to be eroding our country’s leadership position in this area. Herper makes a great point that our society needs to celebrate (and fund) fundamental work. We can’t have any more stories like the one he cites in the Forbes article:

I’m haunted by a story I heard once about a biotech industry lobbyist who went to see a congressman and was told, “You guys don’t do innovation. The iPad. That’s innovative.” [...] As a society, [...] we overlook the work that is actually foundational.

Setting the congressman’s (and possibly the public’s?) ignorance aside, Herper is right. But, innovation in medicine isn’t just hard because medicine is complex. Getting an innovation to this market is just as tough as making the innovation itself. This isn’t good news for aspiring Steves of Medicine like me – we desperately need a Steve Jobs of healthcare/regulatory/public policy before a Steve Jobs of Medicine can emerge.

I just went to the Biomedical Engineering Society annual meeting, and saw a lot of amazing things that scientists and engineers are making. I think it’s just a matter of awareness and communication – more than many other fields, medicine is like football in that it’s a team effort. As long as innovators are in regular contact with policy makers, it’s going to be pretty exciting to see how new technologies can develop a more elegant approach to health care that Steve would have been proud of.

The Audacity of Hope: Biomarkers Edition

The end of the nineteenth century was a turning point in the field of medicine marked by incredible accomplishments, such as the discovery of the smallpox vaccine (Stewart and Devlin) and antibiotics (Landsberg). Medical practitioners no longer used herbs and folk remedies to cure patients, but instead looked towards the emerging developments in biomedical research. In the last one hundred years, progress in biotechnology has enabled the pharmaceutical industry to deliver very sophisticated remedies. Advances have allowed illnesses to become increasingly stratified across multiple populations and disease subtypes. Consequently, some therapies have been developed to target specific, molecularly-defined populations. All of these developments hinge on the hope that we can identify gene or protein biomarkers that can differentiate members of individual populations.

The Problem

Biomarkers, and their corresponding assays and drug targets, were initially developed with the intent of making drugs everybody could use. This philosophy is the idealistic “blockbuster drug” model that is currently entrenched in our industry. The movement for personalized medicine and developing drugs for smaller, more well-defined target populations is catching on, however, the large pharmaceutical companies are reluctant to change their research approaches. At the 2010 Personalized Medicine Conference held at the Harvard Medical School, one of the speakers reported that 94% of companies say they have invested in personalized medicine, but only 10% actually have compounds in late clinical development (phase IIb-IV) (Milne). This suggests that these companies have not made a real commitment to personalized medicine in spite of clear signs that this approach is necessary.

Plato's allegory of the cave. Image taken from Prote Philosophia, Ethica, and Politike (http://ontologicalstatus.blogspot.com/2009/09/platos-allegory-of-cave.html).

In the fourth century BCE philosophical dialogue, The Republic, Plato makes an analogy about people confined to live in a cave. The captives were forced to face a wall for their entire lives, unable to see anything except for the shadows cast on this wall by the objects behind them. Hence, the captives never knew the true nature of the shadows. Their understanding of the world was based on their imagined interpretation of what the shadows represented. Today, Plato might say that the pharmaceutical industry is staring at fuzzy shadows of targeted drugs and calling them blockbusters, when in reality they are missing opportunities to identify promising, targeted drug candidates. In today’s environment, it is increasingly difficult to develop drugs for a broad market since it is harder to prove efficacy for a diverse group—especially with the danger of less controllable risks and side effects for the general population. What will distinguish “good” drug targets in the future is going to be the good biomarkers that lead to them.

While serving as diagnostics for diseases and treatment responses, biomarkers will likely prove that the best treatments are targeted ones. The industry is, however, not yet free from blockbuster drug thinking partly due to a degree of difficulty in identifying effective biomarkers that differentiate sick individuals from healthy ones. Plato might liken the freed captives to the relatively few who pursue personalized medicine approaches, but this liberated group may not get support from the bigger companies who are still shackled to the wall. I argue that organizations investing in personalized medicine are the key to freeing the industry from cave captivity.

As investment in biomarker discovery technology increases, we should establish consistent criteria for biomarker evaluation (Mosedale and Grainger). In spite of the current craze for biomarkers, there are few biomarker candidates that have actually been validated. This hype originates from the lack of such consistent criteria. Designing biomarker discovery and validation studies in a consistent manner would give more legitimacy to biomarker discovery programs, making regulatory approval more facile. Evaluation criteria will also help organize technology development around efficiently obtaining the appropriate information from experiments.

In the post-genomic era, we cave captives have seen “the sun” of incredible amounts of biological information. This information illuminates hundreds of thousands of possible drug targets, but casts a shadow on our collective wall that seems to represent the elusive blockbuster drugs for which our industry yearns. To harness the power of this information, we are in great need of another inflection point. The only way to create it is by facilitating innovation in biotechnology for diagnostics. This challenge is exacerbated by uncertainties in diagnostics reimbursement from health insurance plans, regardless of a diagnostic’s clinical value (Quinn). Comfort is the enemy of greatness, and progress in biomarker discovery and assay development is suffering because of comfort with the prevalent model of drug development.

Escaping the Rut of Comfort

Comfort is the word I choose to describe the industry right now because of the findings of a recent report on the biotechnology industry (Ernst & Young). The industry has a positive cash flow, but long-term concerns are looming. Although the profits of the biotechnology industry have increased over the last year, R&D budgets declined by 20% to leave less “innovation capital” for creating targeted drugs. This problem is exacerbated by the fact that newer, smaller companies are receiving less public and private funding than established companies—according to the Ernst & Young report, 82.6% of funding went to just 20% of US biotechnology companies. Adding to these difficulties, FDA drug approvals are scarcer in the post-Vioxx era.

Funding for biotechnology companies over the past several years. Instead of funding smaller companies with "innovation capital", larger companies are increasingly taking a larger share of the available funding.

Funding challenges aside, scientific instruments are dispensing data as excessively as coins from a winning slot machine in Vegas with, ironically, little reward. Today’s arduous methods of proteomic profiling for blood, urine or tissue produce extremely complicated datasets. Worse yet, these complicated methods are actually simpler than the herculean task of analyzing the data from these experiments! Interpreting this data necessitates a breakthrough in information technology. More importantly, we need a technological breakthrough in high-throughput discovery that supersedes the information technology problem by simplifying the data stream. Genomics has microarrays for high-throughput discovery, but the genome is the tip of the iceberg for many disease etiologies. Unfortunately, protein array technology that can query the entire proteome with a variety of probes is not yet mature for commercial use (Ray et al.), but could one day provide a deep understanding of diseases without data as complicated.

Nevertheless, difficult times inspire great breakthroughs. In the last three years, orphan drugs that target rare diseases have increased from 119 to 175 (Ernst & Young), bringing along with them biomarkers and diagnostics to identify the right patients for their drugs. As industry shifts to developing drugs with particular populations in mind, there is an urgent need to focus on biomarkers and diagnostics. Because the pharmaceutical industry of the future will be focused on targeted drugs, it must invest in developing technology for finding biomarkers. This investment will increase the number of commercially available biomarkers, disease diagnostics and treatment efficacy measures.

A Tough Path to a Sunny Future

In just a little over the course of a century, medicine has changed dramatically. Our industry has gone from understanding relatively nothing about the biology of disease to being able to detect molecular risk factors at an early stage. While we continue to make major progress in biomarkers, we face the need to change the methods industry has become comfortable with in the past few decades of enormous growth.

Science is only part of the solution for improved disease detection and monitoring. Three major things need to change to allow biomarkers to make a much-needed impact in the next era of diagnostics and therapeutics: more innovation capital must go to smaller companies; better reimbursement structures need to be devised for biomarker diagnostics; and collaboration between hospitals, biotechnology, pharmaceutical, academic, and federal institutions must be leveraged for technological breakthroughs. Science and technology will certainly continue to progress, but several important factors lie outside the jurisdiction of science, such as the economics of bringing biomarkers and diagnostics to the market. Specifically, the reimbursement system must be re-built to quickly phase in new diagnostics technologies. Even though the medical diagnostics development and funding process allows only the most valid technologies to reach the market, entire innovative efforts are sometimes wiped out by payer unwillingness. If nobody is going to pay for a diagnostic, the economics make it tough to continue making it (Quinn).

The tangible outcome of these three major changes would be faster discovery of valid biomarkers that illuminate the molecular heterogeneity of diseases. More biomarkers will lead to better diagnostics technologies that will allow us to monitor diseases and treatments more effectively than ever before. Downstream of the assays, good biomarkers will even play a critical role aiding pharmaceutical companies early on in the drug discovery process. With an adjustment of innovation capital, reimbursements, and collaboration partners, the resulting technological innovations will be just what our industry needs for another great inflection point in medicine.

References

  1. Ernst & Young, LLP. Beyond Borders: Global Biotechnology Report 2011. Ed. Gautam Jaggi. 2011. PDF.
  2. Landsberg, H. “Prelude to the Discovery of Penicillin.” Isis 40.3 (1949): 225.
  3. Milne, Cristopher-Paul. “Briefings: Investments in Personalized Medicine.” Personalized Medicine: Impacting Healthcare Conference. Harvard Medical School, Boston, MA. 18 Nov. 2010. Lecture.
  4. Mosedale, David, and Grainger, David. “Biomarkers: Lessons from History | Biomarkerblog.” Total Scientific Ltd. Web. 27 June 2011. <http://www.totalscientific.com/biomarkerblog/?p=55>.
  5. Quinn, Bruce. Coverage and Reimbursement for Molecular Diagnostics: Current Issues and Options. Foley Hoag, LLP, 2009. PDF.
  6. Ray, Sandipan, Mehta, Gunjan, Srivastava, Sanjeeva. “Label-free detection techniques for protein microarrays: Prospects, merits and challenges.” Proteomics 10.4 (2010): 731-748.
  7. Stewart, Alexandra J., and Devlin, Phillip M. “The History of the Smallpox Vaccine.” Journal of Infection 52.5 (2006): 329-34.

 

This was an essay I submitted to the Kelly Scientific Future Scientist Program Scholarship Contest this summer while working at Biogen Idec. That’s why this post is a bit more formal (and more meticulously researched) compared to others on this blog.

Government Funds Technology Entrepreneurs

Wired published this article a few weeks ago on the government’s plan to fund a technology entrepreneurship program at Stanford. The whole point of this $10 million program is to emend how engineers are trained in this country. IMO, this is a great investment by the government—we need technology-based entrepreneurship (and not more “me-too” social media companies claiming to be the next Facebook) to create more jobs in today’s world. I’ve criticized the education system earlier on this blog, but this is great progress. I couldn’t have put it better myself than this:

Think of [this program] as a recognition by the government that entrepreneurship isn’t being taught by MBA programs, but instead has been developed collectively by the culture of Silicon Valley, which has developed a culture of fast development where products are continually upgraded in small — often daily — steps, constant attention to customer needs, and a focus on measuring everything.

Kudos to the government for trying to provide technically competent people a platform through which they can make companies (and jobs) that can improve everybody’s quality of life.

HBR on Managing Health Care

I’m not saying anything new when I say the health care industry in the US is a mess. I like medicine, and I like technology, so I recently picked up the Harvard Business Review on Managing Health Care for inspiration on how I might make my own contribution to the industry. This book is a compilation of 8 case studies on the health care industry. While it’s a bit dated (it came out in 2007), I think it still does justice to the absurdity of things happening in health care.

I come from the technology side, where all of the engineers and scientists are eagerly developing the next breakthrough in medicine. Coming from such an environment, this book was almost a slap in the face of reality for an engineer like me. There are much larger problems than the nitty-gritty of developing a medical technology—problems that can halt the proliferation of even great technologies that work well. Problems aside, I think these case studies give some great solutions to the issues. Here are the three solutions that seemed the most important to me:

  1. Make “focused factories” instead of integrating vertically. “Factories” are organizations that focus on specific diseases (e.g., a kidney center for all cases stemming from renal problems). These factories are more efficient at delivering better, integrated care to patients than systems in which patients are given fragmented care, rife with referrals and inefficient information transfer about the patient/case. Hard data indicates that the costs per case, length of hospital stay, and patient outcomes improve with focused factories.
  2. Switch to consumer-driven healthcare. Consumers aren’t stupid. Before people were given control over their retirement funds, it was feared that the general public wasn’t “sophisticated” enough to make their own decisions about their finances. Some argue that the same fears are undermining efforts to put people in charge of their own medical care. Consumer-driven healthcare would allow people to have flexible spending accounts with which they can pay for care specific to their needs, rather than enrolling in one-size-fits-all managed care plans that effectively make major care decisions for their enrollees.
  3. Enable disruptive and simplifying technologies. What would have happened if a regulatory agency in the 1980s declared, “microprocessors are inferior to logic boards, so all PCs must use logic boards”? The word, “iPhone”, would be nothing more than a typo today. The book’s final case study on disruptive innovations in health care suggests that regulatory bodies are doing something similar to the health care industry today, legally requiring highly specialized clinicians to make diagnoses that could easily be done by “less expensive” people. Basically, doctors are diagnosing the sniffles even though nurses have sufficient training to do the same. Since this case was published, perhaps the increase in physician’s assistants is gradually alleviating this problem. Nevertheless, the authors call for the need to invest in simplifying technologies that reduce the complexity-driven overhead of medicine. If the government, insurance companies, hospitals, doctors, and patients embrace such technologies, then clinicians’ skill levels will be more appropriately matched to the complexity of the medical problems they’re given.

These are the three things I’ll be thinking about while I’m on the drawing board next time. This was a great book that was a quick read, so I highly recommend it. It not only gave me a better (albeit general) understanding of the health care industry, but also a perspective on its future.

Personalized Medicine Conference

I recently attended the Conference on Personalized Medicine: Impacting Healthcare, hosted by Harvard Medical School, Partners Healthcare, and Harvard Business School two weeks ago. I’ve been reading a lot about personalized medicine in scientific journals lately, so I thought this would be a great conference to attend. I learned quite a bit, so I thought it might be useful to share some notes from my experience.

Personalized medicine is a model of medical care that emphasizes the use of information about an individual patient to guide clinical decisions regarding therapeutic or preventative care. The Personalized Medicine conference (hosted by the Harvard Medical School, Partners Healthcare’s Center for Personalized Genetic Medicine, and Harvard Business School) brought together high-profile experts on this topic from industry, government, academia, and healthcare. The conference gave a broad perspective on the progress, limitations, and outlook of personalized medicine in this country, as well as internationally. The head of the Center for Personalized Genetic Medicine, Raju Kucherlapati, says that personalized medicine is in a Dickens-ian “best of times and worst of times” because in spite of great progress, it is still over-promised and under-delivered.

According to Dr. Mark Boguski (Harvard Medical School), 70% of clinical decisions are based on lab tests, either from molecular diagnostics, pathology, or radiology. Because of progress in personalized medicine, he thinks this percentage will continue to increase each year. As an example of progress, Harvard’s Laboratory for Molecular Medicine has added 10 new genetic tests this year (see their page for lists of available genetic tests and their prices).

Progress has also been made in the pharmaceutical industry, as the industry leaders are moving away from the “blockbuster drug” research model in which only drugs that can be prescribed to large populations are developed since they are the most financially rewarding projects. Mikael Dolston, president of worldwide research and development at Pfizer, says his company is moving towards a “precision medicine” drug development model. Precision medicine is an approach that lies somewhere between the personalized medicine and blockbuster drug development methodologies. Precision medicine is the genetic targeting of patient segments, rather than individuals (i.e., they select very well molecularly-defined patient populations) to reduce the costs and size of clinical development and to increase commercial benefit since there is a pronounced treatment effect.

The general sentiment among speakers and the audience was that the science of personalized medicine is advancing faster than the healthcare infrastructure to support it. Cost is one of the major limitations. Life Technologies has reduced the cost of genome sequencing to thousands of dollars and is quickly approaching the $1000 genome. They recently acquired a semiconductor company, Ion Torrent, that may bring the cost of sequencing a genome on a chip to about $500. Still, standard gene tests today cost $1k-$2k per patient. Even when new tests come out, they can be slow to reach the clinic simply because it takes too much time to devise a reimbursement scheme with health insurance companies. IT infrastructure is another area for improvement in clinical and research areas. The need for clinical decision systems that collate the latest changes to clinical guidelines and molecular diagnostics was particularly emphasized by different speakers throughout the conference. Lastly, the conference speakers strongly recommended better molecular medicine education for clinicians, as well as better information dissemination to the general public regarding personalized medicine in order for this approach to succeed.

There were several recurring themes in the conference – precision medicine, collaboration, translational projects, and companion diagnostics. Translational projects and collaboration between industry and academia were mentioned in almost every talk, regardless of the speakers’ affiliation. The executive dean for research at Harvard Medical School decried research in isolation, saying “we need more ideas that don’t just sit around, but that are translated.” He and others went as far as to say that we will never make any progress without collaboration between government, industry, academia, regulators, and clinicians. One of the other most commonly discussed topics was companion diagnostics, the idea of developing drugs alongside a diagnostic test (or even post-market) that can be used to select patients for a particular drug. Companion diagnostics are increasingly being developed – 94% of pharma companies say they’re investing in personalized medicine, but only about 10% actually have compounds in late clinical development with companion diagnostics, meaning there will be more to come. FDA representatives said they will thus help companies integrate genomics into clinical development plans, and their companion diagnostics guidance is forthcoming next month.