Applied Clinical Trials
Successful conduct of hematological malignancy trials requires addressing several unique complexities.
Conducting successful clinical trials in hematological malignancies requires an understanding of a rapidly evolving treatment paradigm that is increasingly nuanced, complex, and patient-directed. Just as the underlying differences in biology and prevalence between blood cancers and solid tumors necessitates differences in treating patients, so, too, do they demand differences in clinical trial expertise and conduct. Sponsors developing hematological oncology therapies must capitalize on the principles and infrastructures shared by solid tumor oncology trials while adapting endpoints, study designs and considering patients' experiences to address the particular challenges related to investigating candidate treatments for blood-based cancers. This article examines the nuances of effectively and successfully conducting hematological oncology clinical trials.
In the United States, someone is diagnosed with a blood cancer every four minutes, and every 10 minutes, these malignancies result in a death.1 Despite this incidence, the three most common forms of blood-based cancers-leukemia, lymphoma, and myeloma-comprised only an estimated 9% of all new cancers and 9% of all cancer deaths in 2013 in the U.S.2 In comparison, the three most common U.S. solid tumor cancers-breast, lung, and colon cancer-together accounted for an estimated 34% of new cancer patients and 43% of cancer deaths in 2013.2 (See Figure 1). Globally, leukemia, multiple myeloma (MM), non-Hodgkin's lymphoma (NHL), and Hodgkin's lymphoma collectively accounted for only 6.5% of all cancer patients in 2012, excluding non-melanoma skin cancer.3
Figure 1. Blood-based cancers comprised only an estimated 9% of all new cancer cases and 9% of all cancer deaths in 2013 in the U.S.
Despite a smaller incidence within the total oncology patient population, the global market for hematological cancer drugs reached an estimated $18.7 billion in 2012, with a projected target of at least $28.8 billion by 2017, equating to a compound annual growth rate of 9%.4
The growing hematological oncology therapy market will be fueled by the success of some of the more than 3,000 medicines in development for cancers.5 Driven in part by the application of new scientific knowledge and technologies to isolate and study the biology of malignant cells, a significant portion of hematological cancer medicines can truly be called novel. Of the 818 hematological oncology investigational projects underway, 627 had the potential to be first-in-class medicines, according to a January 2013 assessment.6
The ongoing research investment in understanding the fundamental biology of hematological malignancies will produce an enhanced and refined understanding of cancer pathologies in general as well as increasing the number of targeted therapies for blood cancers and enabling truly personalized treatments. Therefore, sponsors that ensure careful design and precise execution of their hematological oncology clinical trials will yield data that can inform, and perhaps significantly impact the greater oncology community.
As an example, several years ago, Novella Clinical was approached by a biotech company to help rescue a pair of pivotal Phase III trials for its CXCR4 antagonist in development for hematopoietic stem cell transplantation. The sites participating in the trial had nearly shut down, refusing to submit any further patient data due to the supporting contract research organization (CRO) overwhelming them with nonsensical queries. Simply put, the data and clinical management teams did not understand that, fundamentally, these patients are incredibly sick, and an "abnormal" lab value can be "normal" in this setting. For example, one would expect neutropenia or thrombocytopenia to occur following a transplant, and should not necessarily query lab data showing out of "normal" range white blood cell or platelet counts.
Novella was brought in to leverage its hematological oncology expertise in data management and biostatistics to rebuild the clinical database, clean up a substantial amount of the queries and help perform the analysis to support a new drug application (NDA). Ultimately, Novella turned the trial around and Mozobil(r) (plerixafor) for stem cell mobilization in NHL and MM patients was approved.
A hematological oncology trial is fundamentally different than a solid tumor study. Patient access alone is more difficult, as evidenced by the incidence rates previously mentioned. Researchers should evaluate a hematological oncology study design relative to the feasibility of successfully enrolling and executing the trial and, if deficient, be able to offer modifications or alternative approaches that will lead to successful enrollment. To do so, they first need to understand the pathology, clinical manifestations, and current treatment guidelines of the specific hematological cancer under study to fully comprehend all aspects of the most appropriate trial design. Sponsors and their partners must consider how these cancers influence the selection and precise use of terminology in defining important study parameters, of appropriate trial endpoints, and of data management technologies, as well as the selection and experience of patients.
Defining the disease. The correct clinical research use of terminology associated with hematological cancers requires familiarity and understanding. This defining process can present a notable learning curve for sponsors, trial staff, or partners.
Table 1. A patient-screening comparison.
For example, if a tumor forms in the lung, it is considered lung cancer, which has various subtypes, and if it spreads to other parts of the body, it is still called metastatic lung cancer. In contrast, bone marrow can be the starting location for several distinct cancers, many of which have their own subtypes and some of which can change and develop with time into a new cancer.
Other classification systems define leukemia by the speed at which it develops, either chronic (slow) or acute (more quickly), as well as by the type of white blood cell affected, usually lymphoid or myeloid cells. Similarly, NHL is comprised of a large group of lymphocytic cancers divided into aggressive (fast-growing) and indolent (slow-growing) types that occur from either B-cells or T-cells. Classifying NHL is challenging not only because there are many subtypes, but also because of the methods used to determine the subtype. The World Health Organization (WHO) devised a system that not only uses cell morphology but also includes assessments of cell genetics and surface protein receptors.7
Recently, the most significant advance to clinical trial design, and an early glimmer into the potential of personalized medicine, is the need to identify patients with specific biomarkers before the patient can be considered eligible for study participation. Depending on the frequency of the marker used as entrance criteria, or even whether the marker is tested as standard of care at a given institution, the population of potential patients may be significantly reduced and necessitate a consequently large increase in screenings of patients. For example, adding the criteria that acute myeloid leukemia (AML) cancer patients be positive for a FLT3-internal tandem duplication (ITD) mutation reduces the patient population to just 25% of all de novo (first line) AML patients.8
Understanding the treatment landscape. A sponsor must be knowledgeable of the bigger picture presented by the competitive hematological oncology treatment landscape, including current treatment guidelines and clinical practices. To appropriately drive development strategy in hematological oncology, a sponsor also must understand how to assess an investigational treatment's effect on the underlying disease, not just the patient's symptoms, so as to discern efficacy and safety comparative to current practices. Such evaluations will involve complex assessments using intricate endpoints that may involve counts of white blood cells, neutrophils, myeloblasts, myleocytes, as well as bone marrow measures, the spleen and the liver, among others.
In addition, understanding the investigational therapy is critical. As more and more therapies have the potential to be first-in-class, such as antibody-drug conjugates or mutationally-selective inhibitors, potential clinical trial sites will likely have little to no experience with these new treatments. This knowledge gap can be addressed with appropriate site staff training before and during the trial. Study-specific training often is not just about the therapy but also the processing of complex and sensitive lab samples and the use and measurement of specific targeted endpoints, such as biomarkers.
Determining endpoints. The determination of endpoints differs significantly between solid tumor and hematological oncology. Most solid tumor cancer trials rely on the Response Evaluation Criteria in Solid Tumors (RECIST) to define a participant's improvement/response, worsening/progression or stability. In contrast, the very nature of blood-based cancers requires that treatment trials rely on different measurements to determine treatment-related changes and disease progression, which can add more complexity to trial design, conduct and assessment.
Overall survival (OS) remains the gold standard when evaluating cancer treatment effectiveness. However, progression-free survival (PFS) is the most commonly used surrogate endpoint for trials involving advanced cancers.9 Other progression-related OS surrogate endpoints include disease-free or event-free survival, response rate or objective response rate and time to progression.9
The smaller, single-arm design used in many hematological oncology trials usually precludes using time-to-event endpoints to reliably interpret treatment effects10 and determine OS. Therefore, these trials measure event-free survival, remission rates, duration of response, as well as laboratory measures of biological activity. Major molecular response endpoints also are not uncommon in hematological oncology trials but require great specificity in determining what to measure and the techniques involved, based on the disease under study and whether it is acute or chronic. In the care setting of patients with chronic myeloid leukemia (CML), for example, a polymerase chain reaction (PCR) assay can evaluate molecular responses, namely measures of transcription levels of a specific fusion protein. Such molecular measures are now being adapted to the research setting. For example, entry criteria in CML trials using treatment-free remission as an endpoint require that patients achieve deep, molecular response levels.11
Hematological oncology trials also take advantage of technological developments to measure survival, particularly imaging, to provide greater specificity. For example, the 2011 biologics license application (BLA) submitted to the FDA for brentuximab vedotin was the first to use the agency's response criteria for lymphoma drugs, set forth in 2007,10 which included FDG-PET (18F-fluorodeoxyglucose positron emission tomography) scans in the response assessments. The FDA considered PFS acceptable as an endpoint to confirm clinical benefit because an OS endpoint would not likely occur within a reasonable time frame. The BLA used data from two single-arm studies, both designed to show superiority using PFS as a primary endpoint and OS as a secondary endpoint. The FDA used these data to grant accelerated approval of the biologic for patients with Hodgkin's lymphoma after failure of autologous stem cell transplantation (ASCT) or of at least two prior multi-agent chemotherapy regimens in patients who are not ASCT candidates, and for patients with relapsed systemic anaplastic large-cell lymphoma (sALCL) after failure of at least one prior multi-agent chemotherapy regimen.10
As companies create a development strategy for a compound, the choice of endpoints is very important. Often, sponsors need to strike a balance between FDA-supported endpoints, cost, time, and other endpoints, markers or measures that can quantify or qualify an efficacy signal but may not meet regulatory stringency. Designing trials accurately, understanding appropriate endpoints and measuring response using the correct technology are keys to trial success.
Resourcing study management. Whether in-house or through a partner, clinical research associates (CRAs) need to understand the significance and implications of blood count shifts as well as of transfusions and dosing timing. In the seriously ill patient populations of most hematological oncology trials, certain blood counts are expected to fluctuate because of their disease. For example, anemia in these patients can affect both drug activity and toxicities, so monitoring the anemia and its treatment in trials is important. When hemoglobin counts drop below a certain level, patients require a blood transfusion, which will raise the hemoglobin count. However, the hemoglobin count may also rise after a patient receives a treatment dose. An experienced CRA will know if such fluctuations are prompted by the transfusion, the dosage or the disease state.
Selecting trial sites. Many pharmaceutical and biotechnology companies are pursuing compounds that target hematological malignancies, and combined with an inherently rare patient pool, this naturally causes competition for investigator and institution participation, patient enrollment and key opinion leader relationships. Moreover, the world of hematological oncology specialists is even smaller than that of solid tumor specialists, making competition that much more intense. For a global trial, the rarity of some hematological oncology diseases can result in a lack of knowledgeable investigators or inconsistencies in standards of care from region to region. These variations can limit regional or country-specific options in site selection, which can have implications for trial conduct and regulatory clearance plans.
While the 2012 ASCO National Census of Oncology Practices found more than 70% of all responding practices reported offering a hematological oncology specialty for patients, only 26% of all practice types participated in clinical trials.10 A mere 11.3% of responding practices defined themselves as academic, with either teaching or research activities,12 making it clear that highly knowledgeable and experienced oncologists and hematologists practicing in community settings represent a significant source of referrals for clinical trials. However, at smaller community-based sites, a sponsor may need to offer more significant trial management support.
Consider patients' experiences. Sponsors of hematological oncology studies need to take several patient needs into consideration when recruiting, enrolling, and retaining patients. The trial's educational materials must not only clearly transmit information about potential treatment benefits and risks regarding their disease but also must include the potential impact study participation may have on one's quality of life, such as the number of clinic visits, blood draws, and radiographic studies as common in lymphoma.
Additionally, trial teams must be aware that patients with hematologic cancers have disparate potential viewpoints that set them apart from solid tumor patients and that might impact enrollment and retention strategies. Some patients with lymphoma or leukemia, for example, have a chronic clinical course that extends for years, even decades. These patients are frequently interested in the potential a trial can offer, but want the confidence that an investigative therapy is likely effective and will not interfere with their lifestyle. They usually are less concerned about the pace of a trial and more concerned about invasive procedures, such as repeat biopsies or bone marrow exams, because of their risks.
The opposite is true for patients with acute hematological cancers such as acute leukemia. Time is highly important to them, and they seek immediately available clinical trials and "instant" therapies they perceive as offering the potential for life extension, even if trials involve more invasive interventions, multiple clinical visits, or radiographic studies. These patients find trials requiring long evaluations and limited chances to participate unappealing, but readily accept coming to the clinic frequently to confirm an investigative therapy is working.
The better a hematological oncology trial can address the treatment agendas of individual patients, the more success the study will have in recruitment and completion.
General patient rarity (most of these indications can be classified as orphan) combined with limited treatment specialists and centers, an intense competing trial landscape, and the potential for shifting standards of care both over time and across global regions makes proper site selection that much more critical to a study's success right out of the gate. Disease terminology, classification, and trial endpoints/response assessments are highly complex and can even morph over time, highlighting the need for an active knowledge base in this space when designing the trial and when assigning the study team. Finally, the differences between the potential speed of progression within an indication (e.g., acute vs. chronic leukemia) and the resulting impact of invasive or frequent procedures on a patient's willingness to participate in a given trial should be taken into account when designing the protocol.
Ultimately, the future looks bright for hematological oncology treatment with so many novel drugs and personalized approaches in development; a thorough understanding of the potential challenges in implementing clinical trials in this complex space will help speed the process to bring new therapies to patients in need.
Andrew Zupnick, PhD, is senior director, oncology division, at Novella Clinical Inc., email: email@example.com.
1. The Leukemia & Lymphoma Society, "Facts 2013," August 2013. Page 1. Accessed at http://www.lls.org/#/resourcecenter/freeeducationmaterials/generalcancer/facts
2. "Estimated New Cancer Cases and Deaths for 2013," SEER Cancer Statistics Review 1975-2010, Accessed at http://seer.cancer.gov/csr/1975_2010/browse_csr.php?sectionSEL=1&pageSEL=sect_01_table.01.html
3. World Cancer Research Fund International "Cancer Statistics Worldwide," 2012 data. Accessed at http://www.wcrf.org/cancer_statistics/world_cancer_statistics.php
4. BCC Research, "Global Market For Hematological Disorder Drugs And Diagnostics Expected To Reach Nearly $64.7 Billion By 2017." March 3, 2013. Accessed at http://www.bccresearch.com/pressroom/phm/global-market-hematological-disorder-drugs-diagnostics-expected-reach-$64.7-billion-2017
5. PhRMA, "The Biopharmaceutical Pipeline: Evolving Science, Hope for Patients," January 2013. Page 2. Accessed atwww.phrma.org/sites/default/files/pdf/phrmapipelinereportfinal11713.pdf
6. Long G, Works J. "Innovation in the Biopharmaceutical Pipeline: A Multidimensional View." Analysis Group, January 2013, page 9 and 13. Accessed at www.analysisgroup.com/uploadedFiles/Publishing/Articles/2012_Innovation_in_the_Biopharmaceutical_Pipeline.pdf
7. American Cancer Society, "Non-Hodgkin Lymphoma," Nov. 14, 2013
8. Fathi, A. et al., "Treatment of FLT3-ITD Acute Myeloid Leukemia," American Journal of Blood Research. Accessed atFLT3-ITD acute myeloid leukemia http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3301423/
9. Sherrill, B. et.al. "Review of Meta-Analyses Evaluating Surrogate Endpoints for Overall Survival in Oncology," OncoTargets and Therapy 2012:5 287–296. Accessed at http://www.dovepress.com/review-of-meta-analyses-evaluating-surrogate-endpoints-for-overall-sur-peer-reviewed-article-OTT
10. de Claro, R.A., et. Al., "U.S Food and Drug Administration Approval Summary: Brentuximab Vedotin For the Treatment of Relapsed Hodgkin Lymphoma or Relapsed Systemic Anaplastic Large-Cell Lymphoma," Clin Cancer Res, Nov. 1, 2012 18: 5845. Published Online First Sept. 7, 2012; doi: 10.1158/1078-0432. CCR-12-1803. Accessed at http://clincancerrres.aacrjournals.org/content/18/21/5845.long
11. Mahon, F. and Etienne, G. "Deep Molecular Response in Chronic Myeloid Leukemia: The New Goal of Therapy?" Clin Cancer Res. Published Online First Oct. 29, 2013; doi: 10.1158/1078-0432.CCR-13-1988. Accessed at http://clincancerres.aacrjournals.org/content/early/2013/10/26/1078-0432 . CCR-13-1988.abstract
12. Forte, G.J., et. al. "American Society of Clinical Oncology National Census of Oncology Practices: Preliminary Report," Journal of Oncology Practice, 9 (1), Table 2, p 13. Accessed at http://www.asco.org/sites/www.asco.org/files/oncology_census_2012.pdf
13. Hirsch BR, et. al., "Characteristics of Oncology Clinical Trials: Insights From a Systematic Analysis of ClinicalTrials.gov. JAMA Internal Medicine. 173 (11), June 10, 2013, Published online April 29, 2013.