When it Comes to Cancer Diagnostics, Simpler May Be Better

Cancer is highly complex in nature, so the desire for increasingly more advanced diagnostic tools makes sense. But does complexity always equate to utility? Arnon Chait, CEO and founder of Cleveland Diagnostics, believes in seeking clinically actionable tests that are affordable enough to make a difference at the population level by tackling early detection for some of the most common cancers.

To what extent have we over complicated diagnostics with technological advancements?

You see new smartphones coming out constantly, and it’s human nature to want the next shiny new toy. This concept also applies to science. For example, the next microscope may have better resolution, but we must ask ourselves: will it allow me to do the job better? Is it really necessary to answer a clinically relevant question?

Technology is constantly evolving. From analogue to digital. From image processing and spatial proteomics to machine learning. If you’re a practitioner, what do you choose? Why is one better than the other? Some advancements are needed, and help you do your job better. For example, if you’re a pathologist looking at thousands of images and you’re well trained in machine learning, you can do your job much quicker and with greater attention to detail. But I don’t envy modern pathologists because there are just too many options – it’s overwhelming. In my opinion, new doesn't always translate to better. 

That said, pathology needs to serve the individual; whereas early detection needs to serve the population. But if I give a pathologist a million data points of DNA and RNA, did I do my job better? Or did I just complicate the pathologist’s job?

We cannot forget that cancer represents a very large group of very complex diseases – and we need to respect that complexity to some extent. But my team works with populations, and we recognize the importance of averages. We acknowledge that the full answer is almost always going to elude us and instead look holistically at the disease.

So how do you envision making early detection tests available to the whole world?

Tests have to be affordable – and the healthcare system should be affordable. Even though the United States is “the richest country in the world,” it is the poorest when it comes to the way we distribute health care. There are also complete continents, for example, Africa, that greatly suffer from prostate and breast cancer due to genetics. It’s a huge societal problem. The question is: how do we implement the tests worldwide? The key is to prioritize simplicity and accuracy, which can drive down the cost of distribution, increasing global accessibility in the long term. 

I’m a physicist by training, and we get extra credit in school if we solve the problem in one line rather than one page. On the other hand, biologists are trained in the reductionist approach – that you have to bore into the smallest hole in the middle of the universe to get the answer. If you work in cancer research, good luck to you solving it in a one-line answer! Personally, I look for the path of least resistance – and I believe that’s important in early detection work where a holistic view is essential. We have to step back to see something quite simple. 

And that was our approach to the IsoPSA test that we have developed – we recognize that we don’t need to understand absolutely everything to provide a clinically actionable answer. Indeed, we have a very actionable test that only interrogates one protein (the prostate specific antigen, or PSA) – albeit in a unique way. I understand the science behind it but cannot say how effective it is precisely for every individual – but I can say that, statistically – at a population level – it provides clinically actionable results that signal whether a patient should have a further work-up, such as a prostate biopsy. Remember: simpler is better.