Julie Ross, Ph.D., director of pediatric cancer epidemiology and clinical research at the University of Minnesota, and associate director of population sciences for the University of Minnesota Cancer Center, testified October 26, 2007, before the U.S. Senate Health, Education, Labor and Pensions Committee. Ross' testimony was in support of the Childhood Cancer Research Network (CCRN), which is part of the Conquer Childhood Cancer Act (S. 911), a bill that is currently under review by the Committee. The Conquer Childhood Cancer Act would require the National Institutes of Health to award a grant for the operation of the CCRN, a national nearly population-based registry affiliated with the Children's Oncology Group.

Dr. Ross' testimony:

Thank you for the opportunity to speak today. My name is Julie Ross. I am the Director of Pediatric Cancer Epidemiology and Clinical Research at the University of Minnesota and an Associate Director of the Cancer Center. I am also Chair of the Children's Oncology Group Epidemiology Committee. The Committee's mission is to facilitate the best national research in understanding the causes of childhood cancer.

Over 13,000 children are newly diagnosed with cancer each year in the United States and for several childhood cancers the incidence rates have been increasing. While cure is uppermost in a parent's mind upon hearing a diagnosis of cancer in a child, another common question is WHY MY CHILD?

While we recognize that cancers in children make up only a small fraction of cancers diagnosed in the United States, much of what we know about cancer biology overall has come from studies of childhood cancer. For example, we have learned there are often only two genetic mutations that are needed in order for an eye tumor to develop in a child. This is referred to as the "two-hit" model for cancer. From this model, we now know that some adult cancers can also occur with very few genetic hits. We have also learned of the importance of the in utero period in the development of some cancers following the strong link found between maternal use of diethylstilbestrol during pregnancy and the development of a rare vaginal cancer in daughters. Finally, we are learning much from animal studies, which also point to the importance of the early neonatal period in both childhood and adult cancers.

But we have so much more work to do.

As someone trained in both epidemiology and molecular biology, I work closely with biologists, geneticists, and other basic scientists to try and answer the WHY question.

The Children's Oncology Group, comprised of over 200 hospitals and institutions in the United States and Canada, treats nearly 80% of all childhood cancers diagnosed under the age of 15 years. Nearly all states are represented in the Children's Oncology Group. C.O.G. provides an unparalleled resource for conducting research studies on the causes of childhood cancer. However, this task has become more difficult. While I and several colleagues have been successful obtaining NCI-funded grants to investigate the causes of certain childhood cancers, the resources to conduct such studies are diminishing. Further, it is becomingly increasingly burdensome to conduct these studies on a national level.

Let me give you an example. A research interest that I am quite passionate about is understanding the causes of infant leukemia. This is a very rare childhood cancer; there are only about 130 cases diagnosed in the United States each year. However, infants have a very poor survival compared to other children with leukemia. Further,  there is a genetic abnormality often found in the leukemia cells of infants that is rarely found in older children with leukemia. Importantly, this same genetic abnormality has been found in adults who developed leukemia after being previously treated for cancer with certain chemotherapy drugs. Here is another example of how the biology of adult and childhood cancers is intertwined. We are studying maternal exposures that may mimic the action of these chemotherapy drugs and contribute to the risk of infant leukemia. We have received over $2.5 million in NCI funding to study the causes of infant leukemia. We have some very interesting preliminary clues and this work is internationally recognized for its importance in contributing to our understanding of cancer mechanisms.

However, case enrollments to this study in the Children's Oncology Group have been slow. This is due to the fact that every time a non-therapeutic study such as an epidemiology, survivorship, or similar study is begun within COG, it must get the administrative approval of each institution to enroll patients. This requires considerable resources at the individual institutional level and a strong commitment to participate in non-therapeutic studies. This is often impossible given the more pressing issues of opening clinical trials. The approval process for opening an epidemiology study can sometimes take months or years and some institutions simply do not have the staff time to ever open the study. Thus, some parents of children with cancer are never given the opportunity to participate. Finally, once an institution gains approval to open the study, we often spend a large amount of time and resources trying to locate parents for interview. All of these factors are costly and unproductive and lead to potential critical losses to study enrollment.

This is not the way to do research on childhood cancer. Other countries are doing a much better job by having national research registries. The United States must also come up with a better way.

I have been actively working for many years with others on developing a national nearly population-based registry affiliated with the Children's Oncology Group. This registry is called the Childhood Cancer Research Network or CCRN. We intend to build on and enhance the current registration system in the COG by obtaining informed consent from parents (and patients if age eligible) to allow researchers to access data within weeks of a child's diagnosis. The CCRN also provides for future direct contact of participants by researchers with informed consent. We have piloted the CCRN consent procedures at about 10% of COG institutions over a 5 year period. About 99% of parents agreed to allow access to data, while 96% also agreed to possible future contact. This pilot protocol is considered a major success. It is now time to implement the CCRN at all 200+ COG institutions. The CCRN will serve as a national resource so that we conduct state-of-the art research studies in childhood cancer in the most timely way possible.

Some have asked whether we could use the National Cancer Institutes Surveillance, Epidemiology, and End Results (or SEER) Program to conduct childhood cancer studies. While SEER is an excellent statistical resource that provides information on incidence and trends, it cannot serve as a resource for recruiting study participants, especially for children. Children make up only a very small portion of the SEER program, yielding numbers far too low for rigorous national research studies on the causes of cancer. As an example, the SEER program only captures about 10 infant leukemia cases each year compared to nearly all 130 cases in COG. Further, there is at least a 2-year lag period between final case reporting at a SEER site, making it nearly impossible for an individual researcher to contact patients in a timely way and enroll them in a study.

We are on the cusp of finding some important answers to what causes childhood cancer and also to help ameliorate the potential long-term side effects of treatment. We need the CCRN to best accomplish these goals. I again thank you for this opportunity to present.