Interview with Kannan Natarajan: Drug Development in Rare Diseases - Need for Innovation in Statistical Thinking

November 5, 2019

 stock_lightbulb_bokeh_1000

Cytel is delighted to have Kannan Natarajan speaking at the “Complex Innovative Trial Design Symposium and East User Training” on November 6 in Boston, MA. We got a chance to sit down with Kannan and talk about his career in statistics, the changing role of statisticians, his views on evolving statistical thinking, estimands and relevance of technology in the context of rare diseases.

Cytel: You have been in the industry for 20 years. How did you get involved in statistics?

Kannan Natarajan (KN): I was passionate about mathematics throughout my childhood. Math was the focus during my bachelor’s degree as well. That is when I heard about the Indian Statistical Institute in Kolkata, India. I got myself enrolled and have been involved in statistics ever since then.

Cytel: How has the role of the statistician changed in life sciences during your career?

KN: It has changed to a large degree in a positive manner. In the past, the primary focus was on theory with minimal application as there were computational limitations. There were also a few hindrances in the use of appropriate statistical methods in several areas. Today, statisticians try to understand the science and make the effort in learning more about the problem rather than merely applying a black box tool. Also, the non-statistician scientists appreciate the need for collaboration and input from the statisticians early in the design, and then throughout the trial implementation. We have come a long way and have started recognizing that we have been collecting data for years in multiple different fields and we need to explore this information effectively and hence the advent of the field of data science.

Another tell-tale sign of how statistics has gained tremendous importance can be seen at the present day JSM or Joint Statistical Meetings. In the past, the recruitment center used to be only academia and the government, primarily FDA. But this has changed now, you will see how every company that deals with collection of data, needs statisticians. Even the teaching institutes have shifted their focus from theoretical aspects to application-based study. Applied statistics have become more prevalent and the technology and computational advances have helped greatly in moving forward with some of these changes.

Cytel: Is there increased importance of evolving statistical thinking when working with rare diseases and why?

KN: Certainly, as you will see in my topic, we are improving in the basic science and understanding the pathology of the diseases, particularly in rare diseases and one of the key aspects is that we have a better understanding of the prevalence of some of these rare diseases. Whether it is oncology, a rare tumor, or any other non-oncological indications, there is a problem because of the number of patients that are available for the specific indication that we are trying to assess are far fewer. When that happens, traditional study designs and statistical methods are limited in terms of how you adequately assess the benefit risk of an experimental intervention. This is why there is an increasing need for innovation, and I will be citing some examples of this in my talk to explain how some of these newer methods can actually be more supportive or even replace the traditional methods.

Cytel: Why do you feel there is potential with "estimands" in the context of rare diseases?

KN: As we know, a recently issued draft, ICH E9, is asking clinical trials to clearly describe what are the treatment effects that need to be estimated or an estimand. More importantly, the framework handles what is called intercurrent events that may complicate the estimation of treatment effects. In essence, once you have an intervention of the experimental therapy, all the other intercurrent events that could be relevant during the course of the trial and their impact on the estimation of treatment effect should be categorized. There are also some strategies that are being put forward on how to handle the intercurrent events.

In most cases, rare diseases are life-threatening and life-altering. So, for example, in a situation where a child with Duchenne Muscular Dystrophy is going through gene therapy, as parent of a child with DMD, they would still want to provide additional best intercurrent options for the child during the course of assessment to potentially improve signs and symptoms, regardless of whether the gene therapy is likely to work or not. For rare diseases, the primary estimand that is of interest should have appropriate scientific rigor in order to address the treatment effect considering the intercurrent events. There is also a possibility of looking into biological marker changes, an intermediate or a surrogate endpoint with lesser impact due to the intercurrent events, that could be specifically investigated rather than looking at traditional endpoints such as overall survival. In most cases, especially if you want to look at some of these hard endpoints, you need to have a very long study with a large number of patients, which typically is not feasible in case of rare diseases.

Cytel: What will be the key takeaways for the audience?

KN: A critical aspect of statistics in rare diseases is that we have to find a way to collaborate with non-statisticians. Quite obviously, the key element that we need to understand is the biology and science behind the disease that is being studied. We need to find out its prevalence and then appropriately fit the study design that is necessary to support the benefit-risk assessment of the experimental therapy. That is going to be critical. And, the fact that traditional methods or traditional study designs may not be as appropriate. We need to have a paradigm shift in assessing all sources of information, including augmenting information through Real World Evidence, and not just through a traditional clinical study. We have to find a way to share the data across the industry and academia to find a cure for some of these diseases. Holding onto the data and not being able to share it is not an option if we want to find an appropriate cure for these patients.

At the symposium, I hope to kindle the audience’s interest in innovation and creative thinking. We need to move away from the traditional approach to addressing a problem.

Cytel: Do you see awareness in the industry on the importance of data sharing, as you mentioned?

KN: Luckily for us, people have started recognizing the need for sharing information and data. I feel the awareness started with the understanding the challenge of not being able to go alone but to collaborate more. There is an evolution that is happening across the industry and academia as we are realizing the need for more investment and collaboration in assessing these rare diseases.

Thank you to Kannan Natarajan for his generosity in participating in the blog and we are looking forward to hearing his presentation in Boston.

Kannan Natarajan headshotAbout Kannan Natarajan

Kannan Natarajan is Senior Vice President, Global Head of Biometrics and Data Management at Pfizer.

Kannan has been in the pharmaceutical industry for over 20 years, working across various therapeutic areas and in particular Oncology, over the last 10 years. At Pfizer, Kannan is a member of the Global Product Development leadership team and is the Deputy-Chair of the Clinical Development Strategy Board. Prior to joining Pfizer, Kannan was Senior Vice President and Global Head of Oncology Biometrics and Data Management at Novartis Pharmaceuticals. At Novartis, Kannan was also part of the Oncology Development Leadership Team managing the oncology development portfolio, contributing to the global development strategy and to the approvals of several major drugs. Kannan also served as the co-chair of the Protocol Review Committee, in conjunction with the Head of Development. During Kannan's tenure at Novartis, he was instrumental in managing the growth of Development Operations within India, consisting of multiple line functions within global developments.

Prior to Novartis, Kannan worked at Bristol-Myers Squibb where he served as the Biostatistics and Statistical Programming therapeutic area head for Immunology, Cardiovascular and Metabolics/Endocrinology, supporting sever major global submissions and approvals. Kannan holds a PhD. Degree in Statistics from the University of Florida.

The Challenges in Rare Disease Clinical Trials Symposium

Cytel is proud to be partnering with Alexion to share expertise and ideas surrounding the challenges in rare disease clinical trials. Register today to request a video recording of the presentations and downloadable copies of the speaker's slides.

REGISTER