Tamara P. Miller, MD, MSCE, is determined to improve how side effects of leukemia treatments are reported, thus enabling the true levels of toxicities to be better understood.
The award-winning researcher recently assumed her position as Assistant Professor at Emory University/Children's Healthcare of Atlanta (CHOA) following a successful stint as an attending pediatric oncologist and oncology instructor at Children's Hospital of Philadelphia (CHOP).
Though she has been professionally mobile, this sports-loving, married mother of two daughters counts herself-a Brookline, Mass., transplant-among the Boston Proud. Because her youngest daughter is still a babe-in-arms, Miller says time away from her work consists largely of caring for her family, watching sports-primarily baseball and preferably the Red Sox-and losing herself in a good book. "But that doesn't happen too often with small children," she said through a laugh. "Hopefully, I can get back to reading for fun before too long."
Early Influence
It was actually her early participation in sports that stirred her interest in medicine. "When I was 8, I tore a ligament in my foot while playing soccer and consequently saw a number of orthopedists. That was my first exposure to non-well-child care. It was my first injury and I still remember thinking what those doctors were doing was so interesting-the MRIs, the stuff that is unique to a kid but which now, as a physician, I take for granted. I loved how they took care of me as a patient. It was all so fascinating, so at a young age I knew I wanted to go into medicine."
By the time Miller was in college, she had also identified an interest in working with children. "I had volunteered in settings where I helped younger kids, so I decided on pediatrics. I've never regretted it, because I find children fun to work with," said Miller. "They are hopeful in a way that adults aren't. And they are trusting, funny, and always keep me on my toes."
And then there are the parents.
"When I first did my rotations in med school, I realized a lot of people were put off by pediatrics because they also had to take care of the parents," she explained. "But I found it interesting to take care of both the child and the parent-different 'patients' with different needs. Their wishes and desires may not be the same, so a clinician must approach them differently and give them different levels of understanding as to what is going on medically."
Miller also finds the wide range of ages in pediatrics professionally simulating. "Talking to a 2-year-old is very different from talking to a 15-year-old. That is an exciting, interesting challenge, and it makes me feel really satisfied with the clinical part of my work."
Research Emphasis
Earlier in her career, Miller worked with a group at Boston's Dana-Farber Cancer Institute looking at emotional and neurocognitive functioning in children taking steroids and the changes that occurred when they were on and off the steroids. "I thought it was fascinating how different children could be over the course of the month. That really got me thinking about side effects of cancer therapy. I wondered if we could improve the quality of life for children undergoing intensive cancer therapies in hopes of making the experience as positive as possible for them and their families."
Miller explained, "We are fortunate in pediatric oncology that we have better outcomes than in adults for some cancers. This means that we can start thinking about reducing side effects because we do have better cure rates."
This all begged the question of how best to understand side effects to be able to reduce or prevent them, or, at very least, to prepare young patients and their families for what the child will be experiencing.
Fast-forwarding to her current work, Miller continued, "In order to prepare patients and families as to what to expect, we have to have a better understanding of what the actual rate of side effects are. There has been concern for many years that, while clinical trials have greatly improved outcomes for children with cancer, the side effects that are being reported may not be accurate. I did a chart extraction at 14 hospitals across the country to quantify what the actual rates of these side effects were. I found they are actually higher than what was reported on clinical trials and that, for a large fraction of the toxicities we evaluated, greater than 50 percent of the toxicities were not reported."
Miller said the lack of accurate side effects reporting is not for lack of trying. "The current system uses manual chart review; clinical research associates affiliated with the trials have to read through the medical records and retrospectively report all of the toxicities, and then record the information manually into an electronic system," she explained. "They are relying on what is in the chart and they must search for nearly 800 toxicities. Simply trying to look through and remember all of those, and assign the grading (according to NCI guidelines, with grade 1 being the least toxic up to grade 5 being highest in toxicity) is really a difficult task. It is incredibly time- and labor-intensive, so it is not surprising that toxicities are being missed."
With support from a Damon Runyon-Sohn Pediatric Cancer Fellowship Award, Miller and colleagues are looking at this process and trying to improve the system. "Specifically, we are using electronic medical record data to directly capture the adverse events. We are trying to take out that middleman and just pull the data right from the chart," said Miller with optimism. "The data is already there and recorded for general clinical purposes, so it doesn't have to be added in any certain special way. We are writing algorithms and code to pull the data out of the medical record, proof it, clean it, put it into the right format, and then grade it to match the definitions that are in the published guidelines from the NCI."
Development of algorithms is an intensive undertaking. "We have done laboratory-based toxicities thus far, such as high potassium, and we are just starting to do more complex adverse events, which will take more time and effort," Miller detailed. "Writing the code is relatively complicated, so I think it will take a good 5 years to get all the important toxicities' codes written and tested to ensure the coding is correct and to prove the results are more accurate than the current system of reporting. After that, it will take time to scale it out to a broad range of hospitals and then ultimately to be able to feed it back to the clinical trials and say, 'This is what is happening in hospitals; now we can use this for the actual reporting instead of doing it manually.' It is a 5-10 year plan to really get it up and running. Hopefully I am wrong and it will take less time."
Miller had already started to use the new process at CHOP, and now will test the algorithms and code at CHOA, and then pilot it at additional hospitals. "This is intensive work for us up front, but down the road this could be incredibly time-saving and will free up research associates to focus on other important things," she emphasized.
A Win for Patients
In the end, it will all be a win for patients and their families. "Eventually, we will know the true rates of toxic side effects. We will know how often a patient on a certain therapy gets a kidney injury, or how likely a patient on another therapy is to get pneumonia, etc."
But the first improvement for patients will be found in knowing what to expect, said Miller. "This is helpful for physicians and helpful for patients as they get guidance from their doctors. If you know what the toxicity rates are with standard chemo, you have a better sense of how to compare a new, experimental drug and its side effects to know if it is truly safe or safer than a drug that is already on the market."
In the long-term, Miller hopes her efforts will help reduce morbidity associated with cancer treatments and "... inform decisions that help people determine what they are willing to tolerate to get a better cure rate. Patients will better be able to find the delicate balance between efficacy, cost, and side effects. They can decide if a particular risk is worth taking. While my interest is in pediatric trials, adult clinical trials have the same problems. This could be used anywhere. The new understanding we gain will impact patients of all ages.
"Being able to improve the system of reporting will allow us to use clean, accurate data to answer clinical questions that will help supportive care for patients," she concluded. "In turn, this gets to the main goal of my career: improving quality of life for patients."
Valerie Neff Newitt is a contributing writer.
Spotlight on Young Investigators