Division of Clinical And Translational Research
The Division of Clinical and Translational Research focuses on clinical research in a wide variety of areas including image-guided and intensity-modulated RT, adaptive RT and competing risks analysis. Division Director, Loren Mell MD, serves as Principal Investigator for the Radiation Therapy Oncology Group (RTOG) at UCSD and is the founder of the newly formed International Cervical Cancer Radiotherapy Project.
Clinical And Translational Research Group
Loren Mell, MD
Yun Liang, PhD
Research Program #1
Hematologic Effects and Toxicity Modeling to Augment Pelvic Radiotherapy (HEATMAP)
This is a comprehensive research project to study effects of radiation on pelvic bone marrow (BM), with the ultimate goal to design pelvic radiotherapy (RT) plans that minimize acute and late hematologic complications. Myelosuppression is a common complication of treatment for gynecologic and gastrointestinal malignancies, such as cervical, anal, vulvar, rectal, and many other cancers. Particularly when combined with concurrent chemotherapy, myelosuppression limits treatment intensity and tolerance to adjuvant and salvage chemotherapy regimens. BM injury from radiation is a significant factor contributing to myelosuppression, but the extent to which reducing BM radiation dose can mitigate hematologic toxicity is unknown.
It has long been known that BM functions heterogeneously and is comprised of both active and inactive sub-regions. The locations of active pelvic BM sub-regions in populations of patients undergoing pelvic RT, however, are poorly characterized and understood.
Our research program involves several strategies and novel approaches to optimize BM-sparing RT techniques:
- Identifying active BM sub-regions using functional imaging, such as positron emission tomography (PET) and quantitative magnetic resonance imaging (MRI);
- Developing clinically practical conformal RT plans designed to optimally reduce dose to active BM, using intensity modulated RT (IMRT);
- Mapping critical sub-regions of BM using deformable image registration and high dimensional analytic techniques;
- Developing statistical models of the relationship between radiation dose and hematologic toxicity1-4 ;
- Testing BM-sparing IMRT plans in clinical trials;
- Examining effects of radiation on BM using novel tracers in animal models.
We are seeking to discover the location of critical BM sub-regions using two approaches: “top-down” and “bottom-up”. The “top-down” approach uses data we have collected on patients’ various BM dose distributions and toxicity profiles to trace back the locations of key sub-regions in which higher doses of radiation are most likely to cause toxicity (Figure 1). The approach uses machine learning methods to model dose effects on BM sub-regions.3 Using this approach, we can estimate the clinical effects of sparing these “critical” sub-regions and attempt to understand their functional properties by linking to functional imaging data.
Figure 1 : Image subtraction method showing higher doses to BM in the lumbar spine, upper sacrum, and medial ilium in patients with hematologic toxicity.
The “bottom-up” approach uses functional imaging data to identify and segment key sub--regions based on selected properties, then determine the effect of varying dose levels to these sub-regions. We are using both established quantitative imaging methods, such as 18FDG‑PET and 18FLT-PET, and new methods developed at UCSD, such as fat fraction mapping using T2*-IDEAL (Figure 2),5 and 68Ga-(DTPA)–mannosyl-dextran tracers, to understand the spatial distribution and properties of active bone marrow subregions.
Figure 2 : Axial view of an image-guided functional BM-sparing IMRT plan reducing dose to highly active sub-regions identified on 18FDG-PET and T2*-IDEAL.
Selected HEATMAP Publications
Mell LK, Kochanski J, Roeske JC, et al.
Dosimetric predictors of acute hematologic toxicity in cervical cancer patients treated with concurrent cisplatin and modulated pelvic radiotherapy.
Int J Radiat Oncol Biol Phys. 2006;66:1356-65.
Mell LK, Schomas DA, Salama JK, et al.
Association between bone marrow dosimetric parameters and acute hematologic toxicity in anal cancer patients treated with concurrent chemotherapy and intensity modulated radiation therapy.
Int J Radiat Oncol Biol Phys. 2008;70:1431-7.
Rose BS, Aydogan B, Liang Y, et al.
Normal tissue complication probability modeling of acute hematologic toxicity in cervical cancer patients treated with chemoradiotherapy.
Int J Radiat Oncol Biol Phys. In press.
Liang Y, Messer K, Rose BS, et al.
The impact of bone marrow radiation dose on acute hematologic toxicity in cervical cancer: principal component analysis on high dimensional data.
Int J Radiat Oncol Biol Phys. In press.
Mell LK, Liang Y, Bydder M, et al..
Functional MRI-guided bone marrow-sparing intensity modulated radiotherapy for pelvic malignancies(abstr.)
Int J Radiat Oncol Biol Phys 2009;75:S121.
Research Program #2
Pre-planned On-Line Adaptive Radiotherapy (POLAR)
Most RT plans implemented today are created from static images on a simulation scan obtained prior to initiating the treatment course. However, during fractionated treatment courses lasting many weeks, targets and normal tissues do not always remain static. Considerable morphologic and functional changes have been observed in both target and normal tissues from day to day (“inter-fraction” variation) or even during treatment (“intra-fraction” variation). To compensate for such variations, a planning margin is applied to encompass the range of possible variations. Standard uniform margins are not optimal, however, since changes may be predictable and non-uniform. Adaptive RT refers to approaches that attempt to respond to changes by developing new, optimized treatment plans during the treatment course. This approach requires both a firm understanding and quantification of the extent of such changes, and the development new technological tools to respond to them.
One of the most interesting and complicated disease sites to study is inoperable cervical cancer, because of the wide inter- and intra-fraction variations. Because of constant changes in bladder and rectal filling and the rapid regression of many tumors in response to chemoradiotherapy, the positions of both target and normal tissues can vary substantially each day (Figure 1). Small and large bowels are constantly changing shape and functional changes also occur in bone marrow during treatment (Figure 2). The goals of this project are to model and predict changes to target and normal tissue in patients with inoperable cervical cancer, which would allow us to predict the variety of treatment plans needed. With daily imaging, one could deliver the optimal plan for that day by identifying the best option from a menu of possible plans. Several objectives are:
- Quantify the size of uniform margin required to encompass the target consistently with a desired probability;
- Identify an optimal non-uniform planning margin based on surface landmarking;
- Develop robust shape models to predict the extent of variations in target and normal tissue surfaces;
- Develop a system to generate treatment plan libraries and classify instances of new patient images according to the optimal plan available;
- Test the system clinically to determine its feasibility and efficacy.
Figure 1 : Axial cone beam computed tomography image taken at the end of treatment for a patient with cervical cancer showing discordance between the initial target position (yellow), the position at mid-treatment (blue) and the position at the end (red).
Figure 2 : Axial T2*-IDEAL fat fraction maps showing axial views of the pelvis before treatment, mid-treatment, and post-treatment, showing marked increase in fat content (relative to water), a marker for low cellularity and hematopoietic activity.
Selected POLAR Publications
Tyagi N, Yashar CM, Lewis JH, et al.
Impact of internal organ motion and deformation on target coverage in intact cervical cancer patients undergoing IMRT: a daily cone beam CT study (abstr.)
Int J Radiat Oncol Biol Phys 2008;72:S355.
Lawson JD, Simpson DR, Rose BS, et al.
Adaptive radiotherapy in cervical cancer: dosimetric analysis using daily cone-beam CT (abstr.)
Int J Radiat Oncol Biol Phys 2009;75:S85-6.
Research Program #3
Competing Risks and Endpoints in Clinical Trials (CORECT) Project
Competing risk events are events that preclude or interfere with the observation of primary events of interest. In radiation oncology, investigators are often interested in studying the effect of a treatment on reducing cancer-specific events, such as cancer death or locoregional recurrence. Examples of competing risk events for locoregional recurrence include death from non-cancer causes, second malignancies, and distant metastasis. Competing risks settings are ubiquitous in oncology, including breast, prostate, head/neck, lung, and other cancers. In competing risks settings, precision in estimating the probability of occurrence of primary events of interest is reduced, statistical power is reduced, and inferences based on composite endpoints are prone to error.
The quintessential competing risk event is death from non-cancer causes (i.e., competing mortality). Two strategies to reduce interference from competing risk events are selection and intervention. Intervention involves developing specific treatments or programs designed to reduce competing mortality. Selection involves identifying patients at high risk of competing mortality and designing specific treatments tailored to low and high-risk subgroups. In either case, effective models to risk-stratify patients for competing mortality are needed. Though previous studies have identified prognostic factors for non-cancer mortality, in general, they have not focused on the effects of such factors using multivariable models or using disease recurrence as a competing cause-specific event. The latter approach appears to be effective in stratifying patients into low versus high incidence of competing mortality in head/neck and breast cancer.2-3 We are examining the effectiveness of this approach in prostate cancer as well. The long-term aims of this project are to develop and validate competing mortality risk models in clinical trial research and to assess the impact of competing mortality on efficiency and cost of clinical trials.
Figure 1 : Comparison of outcomes in advanced head and neck cancer according to risk score for competing mortality (death from non-cancer causes).
Selected CORECT Publications
Mell LK, Weichselbaum RR. More on cetuximab in head and neck cancer. New
England Journal of Medicine. 2007; 37:2201-3.
Mell LK, Dignam JJ, Salama JK, et al. Predictors of competing mortality in advanced
head and neck cancer. J Clin Oncol. 2010;28:15-20
Mell LK, Jeong JH, Nichols MA, Polite BN, Weichselbaum RR, Chmura SJ. Predictors of competing mortality in early breast cancer. Cancer. 2010 Aug 24. [Epub ahead of print]
Mell LK, Jeong JH. Pitfalls of Using Composite Primary End Points in the Presence of Competing Risks. J Clin Oncol. 2010 Aug 16. [Epub ahead of print]
Research Program #4
International Cervical Cancer Radiotherapy Project
Significant technological advancements have been made in the radiotherapeutic treatment of women with cervical cancer over the last decade, including the use of intensity modulated radiotherapy (IMRT) and image guided radiotherapy (IGRT). Current work is focused on developing new adaptive RT approaches to further optimize the quality and delivery of RT in these patients.
A barrier to the development and testing of novel technology approaches in this country is the relatively low incidence of cervical cancer. In fact, cervical cancer incidence in the United States has decreased significantly over the last generation. In contrast, cervical cancer remains one of the major causes of cancer death in women worldwide. Consequently, an important strategy to move the field forward is to engage investigators from areas throughout the world where cervical cancer is a major health problem who have the capability of accessing and applying novel technologies in the treatment of these patients. It is hoped that by working together such investigators can test the true value of these technologies in a rapid and efficient manner.
The first clinical trial, a Phase II chemotherapy plus IMRT trial, is currently underdevelopment and will be launched in late 2010 or early 2011. Future trials will focus on other technologies including IGRT and adaptive RT in patients with cervical cancer and other gynecologic malignancies.
To this end, we have established the International Cervical Cancer Radiotherapy Project (ICCRP), a cooperative research initiative that brings together researchers throughout the world with an active interest in radiation technologies for cervical cancer. The group particularly focuses on collaborating with physicians and physicists working in regions with a high incidence of cervical cancer with access to advanced radiation technologies. The first clinical trial, a Phase II trial of IMRT with concurrent cisplatin for stage I-IVA cervical cancer, will be launched in early 2011. For more information, please visit the website: http://cart.ucsd.edu/iccrp. Future trials will focus on other technologies including IGRT and adaptive RT in patients with cervical cancer and other gynecologic malignancies. ICCRP currently consists of investigators from the following institutions:
Artemis Health Institute (Delhi, India)
- Kushagra Katariya, MD
- Subodh C. Pande, MD
- R. Ranga Rao, MD
Instituto do Cancer do Estado de Sao Paulo (Sao Paulo, Brazil)
- Joao Victor Salvajoli, MD
King Chulalongkorn University (Bangkok, Thailand)
- Chonlakiet Khorprasert, MD
- Napapat Amornwichet, MD
Loyola University (Chicago, IL, USA)
- John Roeske, PhD
- Kevin Albuquerque, MD
Peking Union Medical College Hospital (Beijing, China)
- Fu-quan Zhang, MD
- Shuai Sun, MD
Tata Memorial Hospital (Mumbai, India)
- Umesh Mahantshetty, MD
- Shyam Kishore Shrivastava, MD
Website : https://tmc.gov.in/
University of Chicago (Chicago, IL, USA)
- Bulent Aydogan, PhD
- Yasmin Hasan, MD
University of Miami (Miami, FL, USA)
- Aaron Wolfson, MD
- Lorraine Portelance, MD
University of Pittsburgh (Pittsburgh, PA, USA)
- Jong-Hyeon Jeong, PhD
- Sushil Beriwal, MD
University of California San Diego (La Jolla, CA, USA)
- Loren Mell, MD
- Arno Mundt, MD
- Catheryn Yashar, MD
- John Einck, MD
- Steve Jiang, PhD
- Todd Pawlicki, PhD
- Dan Scanderbeg, PhD
- William Song, PhD
- Steve Plaxe, MD
- Michael McHale, MD
- Cheryl Saenz, MD
- Edwin Alvarez, MD
University of Iowa (Iowa City, IA)
University of South Florida Moffitt Cancer Center (Tampa, FL)
Istanbul Bilim University (Istanbul, Turkey)
University Hospital Hradec Králové (Hradec Králové, Czech Republic)
Far Eastern Memorial Hospital (Taipei, Taiwan)