Written by Ι Dr. Ron Korn, Founder, Chairman and Chief Medical Officer at Imaging Endpoints
“I keep dreaming of a future, a future with a long and healthy life, not lived in the shadow of cancer but in the light” – Patrick Swayze
Pancreatic cancer is one of the most lethal malignancies. It is often found late after the disease has spread throughout the body because it doesn’t cause symptoms until the disease has grown large. In fact, up to half of all patients with pancreatic cancer are found during an emergency room visit for sudden onset of jaundice (yellowing of eyes and skin). This usually prompts the treating physician to order a CT scan to look for a pancreatic tumor that may be causing the symptoms.
Once found, it is likely to have spread to other parts of the body such as the liver, lymph nodes and lungs. Cure is very rare, but there is growing optimism as new and better treatment strategies are being developed to shrink the tumors enough so they can be surgically removed. In fact, patients who have pancreatic cancer that is operable (i.e. not metastatic) have a greater chance of being cured than those who have disease that is too widespread for all of the tumor to be removed. Thus, a major strategy amongst oncologists is to make inoperable patients operable.
PharmaCyte Biotech’s Cell-in-a-Box® technology used together with low doses of the anticancer drug ifosfamide provides a new approach to treat this deadly disease by potentially reducing the pancreatic tumor burden enough to offer patients a chance at a surgical cure.
However, knowing whether treatment is working or not is one of the single most critical elements for delivering the right care to patients at the right time. Determining this, of course, is not a simple task. It requires both laboratory and radiology confirmation that a treatment is making a difference. In this article, we will provide a background for the central role of imaging in pancreas cancer treatment, an overview of the role of Imaging Endpoints Research and Core Lab in the upcoming Phase 2b trial, and describe how we have been using advances in radiology science to detect, track and confirm that treatments are working. We will emphasize challenges specific to pancreas cancer imaging and conclude with the unique imaging that will likely be deployed in the upcoming PharmaCyte Phase 2b clinical trial.
Approximately 600,000 American cancer patients emerge uncured from standard-of-care medical, surgical, or radiation treatments each year (1,600 patients every day). For today’s cancer patients, however, there is growing hope. New technologies may enable patients to be matched with targeted treatment options or clinical trials that provide the best chance for survival. To “get it right” requires evidence-based approaches that deliver “the right care to the right cancer patient at the right time,” which can result in measurable improvements in outcomes and a reduction in healthcare costs.
To realize this improvement in outcomes, the right tools must be available to determine if the treatments are effective. At Imaging Endpoints, we have been developing and utilizing advanced imaging approaches to address this need to identify early indications of treatment response. As a result, the use of radiologic imaging has become an essential component for the assessment of treatment activity in any treatment plan involving solid tumors, especially in pancreas cancer.
Imaging provides a non-invasive, low risk approach in assessments of tumor burden prior to therapy and an objective pathway for following treatment response. The definition of tumor response can vary by tumor type, a drug’s mechanism of action (e.g., cytotoxic, cytostatic, targeted, immunologic, etc.), the type of imaging modality used to determine progression (e.g., MRI [Magnetic Resonance Imaging], CT [Computerized Tomography], PET [Positron Emission Tomography]/CT), and/or clinical metrics applied (i.e., tumor markers). Despite the advantages of imaging, routine local radiology practices do not always have the tools or resources to apply these assessments, including RECIST (Response Evaluation In Solid Tumors), the leading assessment of solid tumors in all clinical trials.
RECIST provides a well-defined guideline for evaluating treatment response by evaluating tumor shrinkage though the change in tumor size by measuring tumor diameter during therapy. One of the challenges of using size criteria to determine treatment response is that tumors may not always shrink in response to effective treatment, but instead become stabilized, increase in size due to inflammation and/or undergo a change in their texture. As a result, alternative imaging strategies are sought whenever these situations occur.
Several alternative parameters besides size-based measurements have received attention as potential measures of response including change in tumor volumes rather than diameter, changes in the tumor’s solid nature towards liquefaction (referred to as necrosis or tumor death), changes in tumor biologic activity as measured by PET and change in heterogeneity, hypoxia or perfusion as measured by tumor texture. These alternative methods not only supplement the RECIST assessments, but they can provide vital and more informative assessments of tumor response to help drug developer’s gain a better understanding of their drug and its optimal application. We at Imaging Endpoints have been providing such advanced image analysis for over 100 clinical trials conducted in the U.S. and globally.
The role of the Imaging Core Lab to provide centralized interpretations of scans obtained during clinical trials, is highly sought after by the pharmaceutical and biotech industries to ensure the imaging portions of clinical trials are optimized and standardized across institutions, and help ensure the resulting data meets rigorous regulatory requirements. As an Imaging Research and Core lab, Imaging Endpoints has become a leading provider of centralized radiology interpretations and advanced imaging analysis for the drug development and medical device industry.
In order to provide both sponsor and regulatory agencies confidence in the imaging results, a well-disciplined, strategic approach must be developed early in the clinical trial design to ensure the trial’s objectives can be met. This requires a direct working relationship with the Pharmaceutical or Biotech Company (sponsor), regulatory experts, key opinion leaders, local site investigators, scientists, physicians and other personnel involved in the trial. To facilitate the process of developing the optimal imaging plan for each unique clinical trial, Imaging Endpoints is honored to have a distinguished group of Scientific Advisors and Key Opinion Leaders that provide guidance to enhance the trial design. Amongst our Scientific Advisors are Dan Von Hoff, MD and Manuel Hidalgo, MD, both world-renowned experts in pancreas cancer.
Role of Imaging Endpoints as a Core Lab in Clinical Trials
As we engage in a clinical trial with the sponsor, our corporate philosophy impels us to understand the science and technology underlying a drug’s mechanism of action and how that technology can be successfully translated into the imaging design of the trial. For example, when evaluating the technical and scientific aspects of PharmaCyte’s method of treatment, it is clear that radiologic parameters that measure cytotoxic effects of treatment delivery on tumor response are important, as well as the documentation of Cell-in-a-Box® capsule placement to further explore both local and far field effects of the therapy.
Whether imaging is the primary or secondary objective of a clinical trial, it must be designed with care and incorporated appropriately into the protocol to achieve success. In this manner, our team of experts will work to design the best imaging paradigm possible. Once an imaging paradigm is established for the clinical trial, we then begin to construct the critical documents needed for trial execution. During this study start-up phase, the imaging review charter, imaging manual, project plan and radiology rules are established in conjunction with the sponsor and their team of experts.
This ensemble of activities provides the step-by-step instructions for the conditions of study conduct. There is considerable detail in these documents regarding imaging site assessment and qualification, image acquisition, site training, image scheduling, and secure, real-time transmission of imaging scans to our core lab. In addition, the modalities and methods used for scan acquisition (e.g., CT-Pancreas Protocol, 3D volumetric MRI, FDG (fluorodeoxyglucose) PET/CT, etc.), analysis and radiology interpretation are explicitly described so there is a clear understanding of how, when and why scans should be acquired, and the definition(s) used to declare a radiologic response that are vetted and agreed upon with the sponsor, local physicians (Principal Investigators [PIs]), and regulatory authorities.
Upon trial conduct, Imaging Endpoints works to ensure close communication is maintained between sites, sponsors, PIs and CROs (Contract Research Organizations) for the radiology aspects of the trial. As the imaging studies are completed, they are transmitted electronically and checked real-time for quality per predefined standards by our operational team of leading radiology technologists to ensure they were acquired in the correct manner. Our expert team works with the imaging sites to immediately resolve any identified issues through a disciplined query resolution process.
When all queries are resolved, the images are provided to our expert radiologists for interpretation in an unbiased, blinded manner. This unbiased, blinded process is a crucial step for maintaining the objectivity and veracity of the data. Ongoing monitoring and oversight of the analysis is provided by Imaging Endpoints’ Scientific Affairs and Data Management Departments to ensure data integrity and proper reporting. An advantage provided by our core lab that may be utilized in PharmaCyte’s Phase 2b trial, is real-time interpretations of the images rather than the more typical delayed batch reads at the end of the trial offered by most imaging core labs. This real time evaluation helps to ensure any imaging related problems are identified and corrected real-time, and it is often necessary to make sure there is an objective, uniform determination of response or progression.
For example, it is not uncommon for local radiologists and oncologists to share information during the care of a patient that might bias the local radiologist’s interpretation of the imaging data. This may inappropriately impact the PI’s decision to continue or remove patients from the trial. However, when images are reviewed real-time by the central radiologist expert, this unbiased data can be made available to the local PI under certain conditions. If the central radiology is reviewed at the end of the trial rather than in real-time, then the opportunity can be lost to inform the local investigator in this manner. As a result, patients may be left on trial or taken off trial inappropriately (a process known as informative censoring). Our real-time evaluations may reduce informative censoring that can sometimes mask a treatment’s success.
Finally, at the conclusion of the study or during predefined interval(s) as directed by the sponsor or regulatory agencies, the imaging data is evaluated to determine if the drug has shown promise in the treatment of the disease. Our forensic approach to data analysis extends deep into the changes available on imaging to determine if the treatment is effective or not, and under what conditions the treatment achieves its optimal potential. This “deep-dive” analysis is where Imaging Endpoints excels. Such in-depth analysis and interpretation of the radiology data can make the difference between trial success and failure, and may accelerate the clinical trial and approval process.
Unfortunately, there are too many examples of poor image trial design and execution, where radiology has been incorporated into a clinical trial design as an afterthought with inadequate attention to the radiology technique, design, interpretation and/or analysis. As a result of flawed imaging paradigms, the treatments under review may not meet their endpoints or sponsors must spend extra time and resources rescuing their data to salvage their drug or medical device. Therefore, having a strong imaging partner, like the team at Imaging Endpoints, is one of the key components of trial success.
Use of Imaging to Detect, Track and Confirm Treatment Responses
The basic questions that need to be answered during drug development and approval in any clinical trial are:
- How safe is the treatment?
- What is the optimal dose and route for delivery?
- What are the treatment interactions?
- How effective and beneficial is the treatment?
Each question is addressed through a series of steps, from dose evaluation in Phase 1 studies to identify the optimal dose, to Phase 3 studies that evaluate effects of the treatment on clinical outcome as measured by progression free survival and/or overall survival. Imaging has a key role to play in many of these steps but is most leveraged in determining the efficacy and clinical benefits of treatment. Indeed, tumor shrinkage on imaging has remained the mainstay of efficacy determination over the last quarter century. In fact, tumor size change as evaluated per RECIST criteria is one of the only FDA-acceptable imaging biomarkers to measure treatment efficacy. Thus, almost all clinical trials in solid tumors use RECIST as one of the salient endpoints to judge success.
To ensure that tumor shrinkage tracks with treatment response, two consecutive scans separated by a minimum of four weeks is often used to confirm tumor shrinkage. A reduction in tumor size may be referred to as a “partial response” if tumors shrink by more than 30% from the baseline scan taken just prior to treatment, or a “complete response” if tumors completely disappear. As noted above, one of the challenges of new therapeutics is that they often don’t cause tumors to shrink, but instead disrupt the internal tumor architecture. For these types of responses, newer and more sophisticated analysis must be performed.
At Imaging Endpoints, we use state-of-the-art approaches to measure treatment efficacy providing evaluations far beyond RECIST. One such approach that we have focused upon in our core lab is change in tumor architecture driven by drug treatments.
For this type of analysis, we use a proprietary software analysis technology that interrogates the appearance of tumors on CT/MRI and PET and extracts the individual image elements (digital information on a pixel level that make-up the images of a scan that a physician reviews), and transforms this digital information into histogram-frequency curves. As a result, we can quantitate the extent to which the internal architecture has changed following therapy. This type of imaging biomarker analysis is called quantitative textural analysis or QTA (see Figure 1 below).
Top left panel shows a pancreatic cancer in head of the pancreas (outlined in blue) on CT. Top right panel shows a textural analysis histogram prior to (pre dose) and immediately following (post dose) treatment. Prior to therapy the arterial phase histogram has a much broader shape than the pre contrast histogram suggesting multiple different textural elements are contained in the tumor. Following therapy the two histograms (pre- contrast and arterial phase) converge suggesting early response to therapy. Bottom panels demonstrate a color map of the textural elements of the lesion pre and post therapy with blue and red pixel elements representative of different intensities of signal. The red elements are beginning to merge together suggesting development of necrosis that was not apparent on the CT scans.
Although exploratory, we have begun to use QTA as an early predictor of response (often after only one or two cycles of treatment). Furthermore, there are several publications from our enterprise as well as from leading academic institutions that have shown QTA can predict treatment response and non-invasively extract the major biologic drivers of the tumor from the radiological images. Thus, there is an ever-growing body of evidence demonstrating that texture patterns within tumors can line up with drivers of tumor biology such as tumor heterogeneity, hypoxia (inadequate oxygen supply), proliferation and fibrosis. Such novel methods of advanced imaging analysis are now beginning to provide value in difficult to evaluate tumors such as pancreas cancer.
Preliminary results suggests that when pancreatic tumors respond to treatments such as nab-paclitaxel (Abraxane®) plus gemcitabine or FOLFIRINOX, there is a dramatic alteration of tumor texture compared to non-responding tumors and this change may be related to changes in tumor hypoxia. If confirmed, these new imaging biomarker analysis technologies may be relevant to determining treatment responses for metastatic pancreatic cancers. By serving as surrogate endpoints (i.e., substitutes for long-term clinical endpoints such as survival), these early response biomarkers could hasten the progress of clinical trials.
The opportunities for radiology and nuclear medicine have never been greater, and with these opportunities, comes a responsibility to adapt to the changing trends in imaging sciences. Not all institutions can or will have access to or be trained in all the new imaging technologies. That is why our imaging core lab specializes in adopting and adapting these new technologies into clinical trials. In this way, we are able to see deeper into the subtle changes that can occur during the early stages of treatment response. To advance patient care, we must take the first step toward integrating these advances in imaging with routine care for oncology patients. To this end, the team of experts and institutions who will work together in the upcoming Phase 2b PharmaCyte clinical trial, have a great opportunity to explore the efficacy of the promising new Cell-in-a-Box®-based treatment technology.
Challenges of Pancreatic Cancer Imaging
A critical component of identifying treatment responses in the upcoming Phase 2b PharmaCyte clinical trial rests upon the use of imaging to detect every location of disease within each patient’s body. Equally important is how evidence of disease response will be defined on imaging. Computerized Tomography (CT) is very often the first approach in detection of tumors in patients with pancreatic cancer. With this modality, pancreatic tumors are best visualized on contrast-enhanced CT using what is known in the industry as a “pancreas” protocol.
This pancreatic protocol provides different levels of tumor detectability as the injected contrast “lights up” the lesions when it enters the tumor’s vascular system. Many pancreatic tumors have a reduced blood flow (hypovascular) component best visualized when contrast reaches the tumor veins. If the images on a CT are obtained too early, before the contrast reaches the tumor, it may not be seen. Thus, it is critical that the imaging sites selected to be part of the upcoming clinical trial, be trained by Imaging Endpoints in the optimal protocol for this type of tumor imaging. Magnetic Resonance Imaging (MRI) may also be utilized in imaging pancreatic cancer, particularly for equivocal findings on CT. While there is no particular advantage to using MRI over a contrast-enhanced CT, it may offer better diagnostic detail on cystic lesions, and in the event of a patient’s contraindication to contrast, an MRI is preferred to non-contrast CT images.
A particular issue with imaging pancreatic cancer is related to the infiltrative nature of this tumor. Not only can the tumor cause changes in the native pancreatic tissue, but as the tumor grows it can cause surrounding fibrosis and swelling, leading to a distorted appearance of the anatomy surrounding the pancreas structures, such as the spleen, stomach, duodenum, bile ducts, etc. Sometimes this scar tissue formation can be mistaken for disease rather than tumor. Furthermore, pancreas cancer may spread through the blood stream into other organs like the liver and lungs, as well as spreading to surrounding lymph nodes. As a result, it might be particularly difficult to obtain a comprehensive measurement of tumor burden by CT or MRI.
However, Positron Emission Tomography (PET) is an imaging methodology that can detect the full extent of the cancer based on the cancer’s ability to “take up” radioactive glucose (referred to as FDG), more than normal tissues. As a result, the pancreatic tumors “light up” on PET scans. We have demonstrated that the use of PET scans in monitoring response in metastatic pancreas cancer can be an excellent way to judge the effectiveness of a treatment regimen. And, a PET scan can often be used early, following the start of therapy to rapidly predict treatment response.
Another challenging area in pancreas cancer imaging occurs when imaging is used to determine resectibility of a pancreatic tumor. The condition for surgical resectibility is generally determined by examining the structures surrounding the pancreas tumor invasion (see Figure 2 below). These structures include the major arteries and veins (Superior Mesenteric Artery and Vein, Celiac Axis, Hepatic Artery, Portal Vein), bile duct and lymph nodes.
Figure2: Examples of Pancreas Cancer on CT
Examples of Pancreatic Cancer On Imaging. The top image shows that the tumor (yellow arrow) is located in the head of the pancreas but has not completely spread out to surround the blood vessels (thick yellow arrows). This patient is borderline resectable (operable). The middle picture is from a second patient with tumor that has spread to surround the blood vessels and is not operable because of locally advanced disease. The lower picture shows a large pancreas cancer (outlined in yellow) that has infiltrated the surrounding tissue, blood vessels (circle) and spread to the liver. This is also unresectable because of the widespread nature of the disease.
Surgically resectable generally refers to tumors that are contained entirely within the pancreas; the tumor has not spread along the major arteries and veins or to other organs and the surgeon feels confident that the entire tumor can be removed with clean surgical margins (referred to as R0 resection). Borderline resectable tumors refers to tumors that have grown beyond the pancreas, along the major blood vessels and other tissues. In this condition, the surgeon believes that the tumor can be completely removed not only from the pancreas but striped away from the surrounding blood vessels and other tissues it has invaded so that the margins of the resection will be free of tumor. In locally advanced and metastatic disease, the tumor(s) have extended too far into surrounding structures such that the surgeon would be unable to remove all of the tumor tissue.
The treatment goal for the unresectable group is to try to control the disease and prevent further spread; cure is possible but remote. These patients have traditionally been offered chemotherapy and/or radiation prior to surgery. Usually, disease control is attempted over a 4-6 month period. If disease control can be achieved, then the patient usually undergoes a wait and watch period until the disease begins to grow again. Obviously, the wait and watch time period can be an anxious time for patients and their families, and inevitably the disease returns, sometimes in a more aggressive form.
Consequently, there is a critical unmet need for a maintenance therapy approach to keep the tumor at bay following the initial course of therapy. That is part of the excitement of the upcoming Phase 2b PharmaCyte trial, which will be used in a maintenance therapy situation where very few options exist, and if successful would address a critical unmet need.
With the development of promising therapies, it is now possible to achieve either partial or complete tumor regression. Indeed, some studies are beginning to include the number of patients who go from an unresectable to resectable state as a key clinical trial endpoint. In order to determine this endpoint, scrupulous attention to detail from our central radiologist experts will be required.
At Imaging Endpoints, our expertise in pancreas cancer is well suited for this task. We have achieved various milestones in pancreas cancer, including leadership to demonstrate for the first time, the value of FDG PET in metastatic pancreas cancer response, first to demonstrate remodeling of the extracellular matrix in pancreas cancer with DCE MRI, co-author of the 50th most referenced article in Journal Clinical Oncology 2011 on pancreas cancer response, Core lab for Pancreas Cancer Research Trial Network (PCRT), first to identify a predictive imaging biomarker for the Kras oncogene mutation and intrachromosomal heterogeneity in metastatic pancreas cancer, and first to contribute to the Molecular Medicine Institute for Early Detection of Pancreas Cancer in Arizona. These milestones have fortified our position as being a leading core lab with specific pancreas cancer imaging expertise.
Unique Aspects and Opportunities for PharmaCyte’s Cell-in-a-Box®-based Treatment and Upcoming Phase 2b Clinical Trial
One of the more interesting aspects of the use of Cell-in-a-Box® technology is the methodology of how the treatment gets embedded in or near the tumor. Using interventional radiology approaches through catheter delivery of the capsules (see Figure 3 below), a more directed and controlled manner of product deposition can be achieved.
Figure 3: Interventional Radiology Images demonstrating the insertion placement of the catheter and capsule implantation with subsequent response (Courtesy of PharmaCyte).
Note how the pancreas tumor has shrunken since the delivery of Cell-in-a-Box® capsules to the tumor and subsequent treatment with low doses of ifosfamide.
In this manner the capsules are placed only where the tumor is located, avoiding areas of either normal tissue or errant deposition in other tissues not intended to be treated. Because of this directed approach, the adverse effects from subsequent infusion of ifosfamide may be minimized compared to a more systemic approach.
In addition, directing focal deposition of Cell-in-a-Box® capsules in areas near or around the tumor should subject the tumor to exposures of the drug at higher concentrations compared to a traditional approach of only giving systemic ifosfamide infusions. In this manner, more robust responses to therapy from the ifosfamide should be realized.
Finally, it is possible that delivery of treatment to the primary tumor may affect tumors elsewhere in the body. This is known as the abscopal effect. In the abscopal effect, the shrinking of tumors in different compartments can occur by just treating the primary or major site of disease. The abscopal effect has been initially associated with single-tumor, localized radiation therapy, but may be seen in other types of localized treatments such as intra-tumoral injection of therapeutics.
The use of Cell-in-a-Box®-based therapies has opened up a new way of providing treatment to patients with cancer. One of the greatest areas of an unmet need for new treatments is seen in pancreas cancer because of its aggressiveness. Documenting where the tumors are located within the body and how they have responded to therapy requires the use of radiology scans to obtain a longitudinal picture of tumor shrinkage.
Being able to coordinate and train imaging sites on the optimal role of radiology in pancreas cancer response, and to provide expert centralized interpretations of the scans on a real-time, consistent basis will be a major factor in determining the success of the treatment. Added to this is the novel method of delivering the Cell-in-a-Box® capsules at or near the primary tumor, another radiology driven task. These activities create an opportunity for imaging to become a critical element needed for success. We are honored that PharmaCyte has selected Imaging Endpoints Research and Core lab to help guide this project from beginning to end.