Last month I gave a presentation on cancer medicine and cancer research to a sophisticated group of non-scientists and was asked to predict what cancer medicine would look like in 25 years. This made me think back on a talk I gave in the late 1990s on that very topic. Thankfully, I no longer have the slides I used for that talk! I do recall a couple of items that were a focus of that presentation – one where I missed the mark and another where I was more on target.
I predicted we would find a single (or small number of) common mutations in every cancer type, and that these genetic changes would allow us to identify medicines to treat most cancer types with targeted therapies directed at such changes. My excuse now is that, at that time, we did not understand the complexity of cancer, nor the degree to which cancers that appear identical under the microscope can have very different genetic causes. There have been some important successes in targeted cancer therapy based on genetic analysis but not to the degree I had predicted. On the other hand, I predicted we would figure out how to turn on and off the immune system more specifically, and that this would rejuvenate cancer immunotherapy research. It took awhile, but this has indeed come to pass.
You might think that, given my prior mixed success in trying to predict the future, I would have learned my lesson. No such luck – here I go again. First – the disclaimers. There are two major areas of uncertainty that make it difficult to see into the future. The first is where the science will take us. We are in an amazing era of discovery, and there are surely major advances ahead that will fundamentally change how we think about cancer. The second is how society views and supports research and care across the full spectrum of prevention, early detection, therapy, quality of life and survivorship. Needless to say, no one knows what our political future holds or how society will weigh investments in these areas. While I take responsibility for my predictions, none of the concepts or advances on which I base these predictions are novel ideas. All have been put forth, and are being developed, by many other outstanding investigators and clinicians. Finally, I apologize for the technical language which is difficult to avoid in discussing this topic. Hopefully, the professional lingo will not distract non-researchers and clinicians from grasping the overall points I am trying to make.
I’ll start with cancer prevention. We already know how to reduce the burden of cancer through reducing exposures (tobacco, sun, radon), preventing infections that can lead to cancer (HPV, Hepatitis), and living healthy lifestyles (diet, exercise, weight). Changing behaviors based on this knowledge is another matter. We have seen a steady, if somewhat uneven, reduction in tobacco use over the past few decades that was slow to develop despite irrefutable evidence that tobacco causes cancer. In the same vein, I think we will see slow improvement in other approaches to cancer prevention such as greater acceptance of the HPV vaccine. The science around development of algorithms to influence behavior is advancing rapidly (although, as anyone who follows the news knows, not always in ways that are altruistic). Hopefully, progress in the science of behavior will be used in the years ahead to encourage healthy behaviors. Such progress will only be possible if we also have a committed public health community that understands the long arc necessary for healthy behaviors to be broadly adopted.
New science will provide new approaches to cancer prevention as well. The microbiome is gaining increasing attention, and we are just starting to figure out the microbiome impacts on health, disease and therapy. Modifying the microbiome (through diet, medications or even introduction of select organisms or microbiological ecosystems) could well be used in the future to reduce the risk of cancer for select individuals or populations.
Some of the most dramatic progress is likely to come through personalization of early detection, and application of less invasive, molecular tests to detect premalignant lesions and early stage cancers. New technologies such as liquid biopsies and stool testing for shed cancer-associated genes and genes associated with pre-malignant lesions will increasingly provide non-invasive ways to identify previously invisible abnormalities. What is needed now are robust algorithms that can help us interpret data on the possible presence of early stage cancer that comes from various sources of information. These algorithms will not only be necessary to assess the molecular information, but also to use that information to identify the best approach to management of each patient. False positive results (tests that are positive when, in fact, there is no disease) emerging from more sensitive new approaches could have immense cost with respect to both patient anxiety and dollars. Understanding the impact of false positives will be increasingly important as algorithms become more complex and testing becomes more sensitive. A vital component of such personalization of early detection will be identification of patients with pre-malignant or early stage cancers who do NOT need invasive or aggressive therapy. Despite these challenges, I am confident our approaches to early detection will be more individualized, more sensitive and more accurate than those we are using today.
Progress in prevention and early detection will also be highly dependent on the social and political choices we make. Today, incentives for payers and health care systems drive them to focus on short term costs, not long-term outcomes. If incentives for investing in the long term health of our communities remain relatively limited, then investment in research focused on cancer prevention and early detection, and the resulting progress, is likely to be slower. However, if society becomes more dedicated to the long term health and well-being of those we serve, then I see more attention, greater investment and greater progress in our ability to prevent and detect early stage cancer.
When thinking about the future of cancer therapy, the lessons of targeted therapy directed towards specific genetic changes in cancer are important to consider. When I made my first set of predictions a couple of decades ago, we were all incredibly excited about the success of Gleevec for treatment of chronic myeloid leukemia (CML). We were confident many additional targeted therapies would follow and rapidly change the face of cancer therapy. While there have been additional important breakthroughs in targeted cancer therapy, success has been slower to emerge than many (including I) predicted. This is in large part because of the unexpected genetic complexity and heterogeneity of cancer.
Front and center in the area of cancer therapy these days is cancer immunotherapy. Given the experience with targeted therapy, a central question in cancer immunotherapy is whether recent progress will continue to accelerate, or whether we have already identified the “low hanging fruit” and future progress will be slower. I will go out on a limb and predict we are just scratching the surface with cancer immunotherapy. I believe checkpoint blockade is just the first of a series of major advances in our understanding of cancer and the immune system, and future discoveries will have major clinical implications. Additional advances will include modification of the tumor microenvironment to improve antigen presentation and T cell activation in tumors. Molecular and computational advances will allow us to predict which “neoantigens” are most likely to be capable of inducing a clinically meaningful anti-cancer immune response. The result will be successful immunotherapy for cancers that currently do not respond to current approaches. I will go so far as to say that, in the decades ahead, immunotherapy will replace cytotoxic chemotherapy as the primary approach used to treat cancer. Cytotoxic chemotherapy and radiation will still have a role, but in most cases, will be combined with immunotherapy.
I have a similar level of optimism about Chimeric Antigen Receptor T cell (CAR-T) therapy. This exciting advance in cancer immunotherapy involves genetically reprogramming T cells to attack cancer cells. Using different types of engineered T cells should allow us to develop CAR-T that target not only molecules on the surface of malignant cells, but also intracellular tumor associated antigens. This will markedly expand the types of cancer that can be treated with CAR-T. Key to broadening success of CAR-T will be the ability to tune up or down both the intensity and timing of the anti-tumor CAR-T response and so make such CAR-T more effective with side effects more controllable. If we are able to do so successfully, CAR-T and related methods will replace current approaches to stem cell transplantation that involves cleaning out the bone marrow with strong chemotherapy, then waiting for the normal marrow to regrow.
Despite the scientific and clinical excitement over progress in cancer immunotherapy, advances will only have a meaningful impact on the burden of cancer if we figure out how to make them affordable. The current cost structure for such treatments is not sustainable. Combination cancer immunotherapy treatments are going to become more common, and this will further exacerbate the cost problem. As a society, we have not yet begun to try and address the issue of cost and of value for cancer therapy. We will need to do so in the years ahead if our patients are to obtain maximal benefit from ongoing scientific advances.
Finally, with more effective therapy will come a greater need to focus on survivorship and quality of life. Research in these areas is only just beginning, and support for such efforts is impeded by some of the same societal pressures, and lack of a profit motive, outlined above that have limited progress in cancer prevention and early detection. Somehow, we need to convince our political leaders that these efforts are important. This need should become clearer as the number of cancer survivors increases. We are making steady, if slow, progress in incorporating palliative and supportive care into cancer treatment, and I suspect this gradual progress will continue.
So there you have it. My predictions for where cancer medicine will be in 25 years. Progress – albeit slower than we would like – in cancer prevention, survivorship and quality of life. A paradigm shift in early detection based on new molecular approaches, and the emergence of cancer immunotherapy as a central component of cancer treatment. A need to more rationally address the financial aspects of cancer medicine across the full spectrum of cancer prevention, early detection, therapy, quality of life and survivorship in order to optimize our ability to reduce the burden of cancer.
Perhaps someone will keep this blog, and give me a call (if I am still around) in the retirement home in May, 2043 to let me know how I did…