Clinical Trial Endpoints for Oncology Studies

Innovative clinical trial endpoints are intended to evaluate potential new cancer treatments. While early phase conventional clinical studies demonstrate more tangible endpoints such as tumor shrinkage, later phase endpoints may evaluate prolonged survival and quality of life. Endpoints serve as a standard for measuring clinical research outcomes.

Laurie Strawn, PhD, presented the six most common oncology clinical research endpoints at the RAPS Annual conference in San Jose.¹ Strawn explained that primary endpoints must provide a valid and reproducible measure of clinical benefit in the patient population. Some examples may be defined as type of tumor, line of treatment, or previous treatment. “Endpoints need to be defined prospectively in the protocol and statistical analysis plan,” Strawn added.¹ The methods to measure these endpoints must be standardized and the clinical effect must be meaningful in addition to being statistically significant. These cancer clinical trial endpoints are outlined in Table 1.

Overall survival
Overall survival is the gold standard for cancer clinical research. It is defined as the time a patient begins treatment until death from any cause. Overall survival is measured in the intent-to-treat population and is considered the most reliable cancer endpoint.² When survival is expected to be limited and there is no opportunity for further treatment, overall survival is the preferred endpoint. Demonstration of statistically significant improvement in overall survival is clinically significant when the toxicity profile is acceptable.² The endpoint is easy to evaluate without reliance on tumor measurements.

The disadvantage of relying on overall survival is the possibility that it may require longer and larger studies in order to prove efficacy. Any subsequent treatments for the patient may have an effect on overall survival. Many clinical studies use a control arm and an experimental arm. The control arm can be the current best treatment known to be prescribed for that patient group while the experimental arm is the study drug. When studies are designed to permit the control arm to cross over and switch with the treatment group at the time of disease progression there is a greater potential to obscure a survival benefit.

Overall survival controversy
There was a recent controversy concerning overall survival endpoints in a cancer clinical trial for a melanoma treatment that shrinks tumors in some patients, for a limited time.³ The clinical endpoint for this trial is overall survival and it was not known which patients would live longer. Two cousins were enrolled in the same cancer clinical trial for the treatment of melanoma. Both cousins were randomized by computer and one cousin was assigned the study drug and the other received the traditionally prescribed treatment for melanoma without the study drug. The cousin receiving the study drug was improving while the cousin who did not was doing poorly.

This trial was operating by traditional regulatory standards for approval of a new drug except the new drug was not a toxic chemotherapy. The experimental drug is a known cancer enzyme target that works on melanoma patients with a specific mutation (V600E mutation). Occasionally Phase III clinical studies that help patients with no other options have controversies when the experimental drug has competing results—especially when the overall survival is used as a primary endpoint. The FDA has compassionate use policy for providing excess to promising experimental drugs.

Progression free survival
Progression free survival is defined as the time when a patient is treated to the first documentation of objective disease of death of the patient by any cause. This endpoint provides measurements of tumor growth prior to the determination of survival.² Progression free survival is a useful tool when studying patients with high unmet medical needs. There is less confounding by crossover or subsequent treatments because progress must be documented prior to the beginning of subsequent treatments.¹ Progression free survival data analysis is confounded when patients miss scheduled exams or drop out of the study. Another disadvantage of progression free survival is that the precise timing of the interruption of cancer progression is difficult to determine. This may lead to investigator bias. Sometimes the underlying disease obscures the ability to measure the cancer progression.

Disease free survival
Disease free survival is defined as the time the patient is treated until the first documentation of relapse or death. Disease free survival is also referred to as relapse free survival. It is used as an endpoint following surgery or radiation therapy. Disease free survival is an important endpoint when survival is prolonged, making overall survival endpoints impractical.² Breast cancer hormone therapy treatments are frequently measured with this endpoint because they carry minimum side effects with favorable risk benefit relationships. This is particularly advantageous after surgical removal of tumors.

Disease free survival can be difficult to determine if the patient dies prior to documentation of tumor progression. It does however allow for small shorter clinical studies. Investigator bias can be minimized by blinding patients and investigators to the placebo or treatment arms.²

Objective response rate
Objective response rate is when a proportion of patients exhibit tumor size reduction of a predetermined amount for a minimum time period.² The response duration can be measured from the initial response time to the documented tumor progression.² This can also be evaluated by the sum total of partial response plus the complete response. Objective response rate does not include stable disease or necrosis without tumor shrinkage. Only patients with measurable disease would qualify for enrollment with this endpoint. It is useful for patients requiring subsequent treatments and there is often the benefit of small shorter studies. On the other hand, it is not a precise measure of drug activity and it is subject to investigator bias.

Durable complete response
Durable complete response is the durable complete regression of the progressing disease. It is very useful when no alternative treatments are available to the patient. When single arm trial is employed, the natural history of the disease without therapy must be well characterized.¹ While this endpoint only benefits a small subset of patients it is an established surrogate for quality of life and survival.

Patient reported outcomes
The rarely used patient reported outcomes endpoint may show a correlation between symptom improvement and treatment benefits. Although the precise definition of meaningful change in symptoms is difficult to standardize, it can enhance the objective response rate findings. Patient reported outcome measurements are also used in post-marketing studies.

Secondary endpoints
The six endpoints listed above are the most common oncology clinical study endpoints. Clinical trial investigators often tailor the definitions based on their specific study. The actual endpoint and corresponding statistical analysis is clearly defined prior to the onset of the trial. Oncology studies are known to have a primary endpoint and secondary endpoints to support the trial outcome.

Improving endpoints
Traditional chemotherapeutic agents act directly on tumors while cancer immunotherapies act on the immune system. Immunotherapies are often associated with delayed clinical effects demonstrated by no prolongation in time to progression, but longer survival post progression.⁴ Immunotherapy oncology treatment can induce novel patterns of antitumor responses distinct from those of chemotherapy and require modifying the tumor burden measurements.⁴ Immunotherapies work on cancer cells by delaying their growth and not by directly shrinking the tumor cell.

According to Albert Deisseroth, MD, PhD, who spoke at the RAPS Annual conference in San Jose, “We are entering into an era where shrinkage of cancer tumors is no longer required for treatment to be successful.” He also added that, “new investigational oncology treatments are able to extinguish cancer signal transduction pathways within the cell instead of killing the cell."⁵ There is a need for surrogate endpoints to characterize and evaluate new investigational drugs.

Surrogate endpoints
Surrogate endpoints are biomarkers substituted for clinical endpoints.⁶ Biomarkers are measured and evaluated as an indication of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention.⁶ Biomarkers are much easier data points to collect than clinical endpoints. Measuring blood pressure would be the biomarker. Surrogate endpoints need to be specific to both the tumor type and the investigational drug. Surrogate endpoints in cancer studies need to be validated and reinforced with traditional endpoints.

Validating new endpoints is necessary since the therapies are targeting very specific biochemical pathways within the tumor cell. Deisseroth stated, “A clinical trial may occur where the surrogate endpoint does not predict the classically accepted endpoint."⁵ When developing new endpoints, we must consider its biochemical mechanism of action, the type of cancer, and the patient population. According to Deisseroth, “The standard for adopting surrogates is going to be very high and the data must be backed up by more acceptable endpoints."⁵

Adaptive clinical trials
Adaptive clinical trial designs are becoming more popular in oncology clinical studies and are gaining acceptance with the FDA. Adaptive designs are multistage studies that use accumulating data to determine how to modify specific aspects of the study without destroying the validity and integrity of the clinical study.⁷ The purpose of a multistage adaptive clinical trial design is to accumulate information and data on the outcome of the results in earlier stages of the clinical study. Adaptive clinical studies can compare several drugs or treatment doses and through the course of the clinical study focus on the remaining two that appear to be most beneficial to the patient. Adaptive clinical trials are best suited for advanced and metastatic disease where sufficient events are obtained before enrollment closes.⁸ Adaptive clinical trials evaluating biomarker driven oncology studies have been successfully used in Investigational New Drug Applications.

The parameters of an adaptive clinical trial may be modified but they must be specified prior to starting the clinical trial. The following clinical trial parameters may be modified if they are clearly specified prior to starting the clinical study: number of patients enrolled, study duration, endpoint selection, treatment duration, patient population, number of treatments, number or interim analyses, and the hypothesis.⁷ In adaptive clinical trial design, the goal is to target the optimum safe dose.⁷ The optimum safe dose is almost always smaller that the maximum tolerated dose.⁷ Adaptive clinical studies are able to address more research questions in one clinical trial without having to re-enroll patients. They do however require a considerable increase in planning and resources especially for the statistical analysis.

Technology is driving the regulations. The process of developing and validating acceptable new clinical endpoints will require time to collect and evaluate the data. The focus on targeting therapeutics to changes in cancer cells creates more therapeutic options.

References

1. Laurie Strawn, PhD, Sr. Director, Worldwide Regulatory Strategy, Global Regulatory Portfolio Leader for Genitourinary and Gastrointestinal Tumor Strategy Teams at Pfizer Inc., "Oncology Drug Development Session: Endpoints for Cancer Drug Development and Approval," presented at the 2010 Regulatory Affairs Professional Society Conference & Exhibition, San Jose, California.
2. Food and Drug Administration, Guidance for Industry Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics, (FDA, Rockville, MD, 2007), http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM071590.pdf.
3. Amy Harmon, "New Drugs Stir Debate on Rules of Clinical Trials," New York Times, (18 September, 2010), http://www.nytimes.com/2010/09/19/health/research/19trial.html.
4. A., Eggermont, S. Janetzki, F. Hodi, R. Ibrahim, A. Anderson, R. Humphrey, B. Blumenstein, L. Old, and J. Wolchok, "Improved Endpoints for Cancer Immunotherapy Trials," Journal of the National Cancer Institute, 102 (18), (2010).
5. Albert Deisseroth, MD, PhD, Division of Hematology Products, Center for Drug Evaluation Research, FDA, "Oncology Drug Development Session: Regulatory Flexibility in Oncology Drug Development," presented at the 2010 Regulatory Affairs Professional Society Conference & Exhibition, San Jose, California.
6. Biomarkers and Surrogate Endpoints in Clinical Research: Definitions and Conceptual Model, NIH Definitions Working Group, 1–9, Elsevier: 2000.
   
7. Vlad Dragalin, PhD, "Adaptive Oncology Clinical Trials," presented at the International Molecular Med Tri-Con 2012, San Francisco, California.
   
8. Cyrus Mehta, PhD, "Adaptive Designs for Phase III Oncology Trials: VALOR Trial and Extension," presented at the Molecular Medicine Tri-Conference 2012, San Francisco, California.