Clinical trials endpoints

The development of novel cancer treatments especially relies on clinical trials and the utilization of relevant clinical endpoints. In trials with cancer patients, it is also important to account for the different cancer types, histological subgroups, and life expectancy [ 8 ].

For example, while clinical endpoints were traditionally reserved to evaluate therapies during phase III clinical trials, their use in early-phase clinical trials has become more frequent with the development of novel immunotherapies [ 9 ].

This paper will explore the benefits and drawbacks of using different clinical endpoints in clinical trials that assess the efficacy of new cancer treatments.

Overall survival OS is defined as the time from randomization to death [ 10 ]. Any patients lost to follow up or still alive at the time of evaluation are censored [ 11 ]. Since the goal of cancer treatment is generally to extend survival, OS is often referred to as the gold standard endpoint in oncology clinical trials [ 6 ]. OS is a patient- centered endpoint that is easy to measure; it is definite since the final time point is death [ 12 ].

Moreover, OS is objective, and researcher bias is unlikely to take place [ 13 , 14 ]. While OS remains the preferred clinical endpoint in oncology clinical trials, it has some drawbacks. Primarily, the expectation of long-term patient follow-up indicates a larger patient population is required and the study will require more financial support [ 15 ].

Additionally, OS has limited use in diseases that are slowly progressing and have an expected long-term survival. In these cases, OS may be influenced by treatment in further steps, sequential use of other agents, and cross-over treatments, making it difficult to attribute the clinical endpoint to a specific medical intervention [ 4 ].

Additionally, as a primary clinical endpoint, OS can also be influenced by non-cancer deaths since the endpoint is defined as time from randomization to death of any cause [ 10 ].

Progression free survival PFS is defined as the time from randomization until first evidence of disease progression or death [ 10 ]. PFS is measured by censoring and patients who are still alive at the time of evaluation or those who were lost to follow up [ 14 ]. PFS is a popular surrogate endpoints since fewer patients are needed to obtain the data that becomes available early in the trial [ 14 ].

These factors often lead to early completion of clinical trials and reduced associated costs [ 13 ]. PFS also provides the benefit of objective evaluation without being influenced by subsequent therapies or cross-overs [ 4 ].

PFS is an attractive choice in clinical endpoint for its direct information with regards to drug activity and their rapid turnaround in data compared to OS. Additionally, PFS has drawn more attention as a clinical endpoint for its ability to assess treatment paradigms that include multi-stage therapies.

Where OS fails to assess the short-term, incremental changes of each round of treatment, PFS can [ 16 ]. Time to progression TTP is defined as the time from randomization until first evidence of disease progression [ 15 ]. Since PFS and TTP are similar, it is important for studies to clarify what is meant by evidence of disease progression. In advanced breast cancer, some investigators use PFS and TTP interchangeably, potentially leading to confusion when comparing the outcomes of various trials [ 17 ].

Meanwhile, studies have used TTP to evaluate aggressive therapies for advanced non-small cell lung cancer, however, its use as a surrogate marker is not definitive and it should be avoided as a primary endpoint [ 18 ].

This variant of TTP compares tumor growth both on and off treatment, serving as an intra-patient control for natural tumor growth rate [ 19 ]. Disease free survival DFS is defined as the time from randomization until evidence of disease recurrence [ 4 ]. It is closely related to duration of response DOR : the length of time a tumor will respond to treatment without growing or metastasizing [ 18 ].

While PFS is used as a clinical endpoint for treatments that are used to manage more advanced, metastatic malignancies, DFS is used as a clinical endpoint for adjuvant treatments after definitive surgery or radiotherapy. For this reason, DFS is often regarded as an important end-point for cancers with a prolonged OS [ 20 ]. The value of DFS has been contended by experts in the field for a number of reasons. Moreover, questions of the validity of an incidental finding of cancer regardless of symptoms in such patients continue to be controversial [ 20 ].

For these reasons, it is increasingly important that recurrence be defined when utilizing DFS as a clinical endpoint. Despite its draw backs, DFS has been used as a strong surrogate endpoint for OS in clinical trials for stage III colon cancer, in an adjuvant setting in lung cancer, and in breast cancer [ 21 - 23 ].

Event-free survival EFS is defined as the time from randomization to an event which may include disease progression, discontinuation of the treatment for any reason, or death [ 24 ]. EFS is a surrogate endpoint that can be used in the place of a primary endpoint, such as OS, to reduce sample size, costs, and duration of follow-up [ 25 ]. Since EFS is a surrogate endpoint, it needs to be validated for each unique tumor type, treatment, and stage of disease. One of the benefits of using EFS over OS in cases of AML is that it is not dependent on therapy given after failure to reach, or relapse from, remission; this quality means EFS provides a more direct assessment of therapy-based benefit during treatment induction [ 25 ].

Reasons for prematurely discontinuing treatment can include cancer progression but also adverse events, patient choice, or death. Unlike other clinical endpoints, TTF is regularly used for regulatory drug approval since it does not directly measure treatment efficacy [ 27 , 28 ].

When TTF is used as a primary endpoint, secondary endpoints are strategically chosen to explore the portion of patients that discontinued chemotherapy due to disease progression compared to other reasons [ 29 ]. It is also important to consider the age of patients enrolled in the trial when using TTF as a clinical endpoint. Older patients experience higher rates of adverse events during clinical trials; this difference could impact TTF due to adverse events and should be considered during study design [ 29 ].

Time to next treatment TTNT is defined as the time from initiating treatment to initiating the next line of therapy [ 30 ]. In low grade, incurable diseases, TTNT is a meaningful endpoint for patients who will require many therapeutic interventions to extend survival.

Unlike most disease-related endpoints, TTNT includes the time course of treatment tolerability and patient compliance [ 30 ].

TTNT continues to emerge as a measure of duration of treatment efficacy, specifically in primary cutaneous T-cell lymphomas [ 31 ]. Despite its usefulness, TTNT is a surrogate marker for duration of clinical benefit and requires validation before serving as a standalone marker to assess treatment efficacy. Duration of clinical benefit DoCB is defined as the time from randomization to disease progression or death in patients who achieve complete response, partial response, or stable disease for 24 weeks or longer [ 32 ].

It is a primary endpoint that is used in clinical trials in which disease stabilization in order to prolong survival is the primary goal. Duration of response DoR is defined as the time from randomization to disease progression or death in patients who achieve complete or partial response [ 32 ].

It is closely related to DFS and measures how long a patient will respond to treatment without tumor growth or metastasis. DoR is useful in assessing treatments that promise durable response and delay disease progression as opposed to treatments that provide a temporary remission without lasting benefit [ 32 ].

Objective response rate ORR is a measure of how a specific treatment impacts tumor burden in a patient with a history of solid tumors [ 33 ]. It is defined as the proportion of patients that respond either partially or fully to therapy [ 34 ].

ORR is a good measure of anti-tumor activity, and there are many different evaluation criteria that have been used to assess ORR in the past [ 4 ]. ORR was primarily assessed by tumor size and total tumor load as found via different anatomic imaging modalities [ 35 ]. Despite its favorability for nearly two decades, the WHO criteria fell out of favor for its interobserver variability of the number of lesions and the selection of measurable targets [ 36 ].

In its most recent version, RECIST selects target lesions by size and defines them as representative lesions of all involved organs. A maximum of 2 lesions per organ and 5 in total are considered baseline target lesions while all other lesions are non-target [ 33 ].

Primarily, ORR fails to capture patients with a stable disease, and it does not differentiate patients with complete response from those with partial response [ 9 ]. For now, ORR provides the most utility in trials evaluating neoadjuvant therapies, especially those in breast cancer patients [ 4 ]. Complete response CR is defined as the lack of detectable evidence of tumor [ 37 ]. Imaging studies and histopathology are used to measure CR which can be used as a surrogate or primary endpoint depending on the specific disease or context of use [ 37 ].

For example: CR in the setting of multiple myeloma therapy has proven to be clinically relevant as it conveys a survival advantage associated with improved OS and prolonged EFS in specific treatment studies [ 38 , 39 ]. Pathologic complete response pCR has most oftenbeen used as a surrogate marker in breast cancer and is defined as the absence of residual invasive cancer upon evaluation of the resected breast tissue and regional lymph nodes [ 40 ].

In studies following patients after neoadjuvant chemotherapy for breast cancer, patients who achieved pCR were associated with improved OS and EFS [ 42 , 43 ]. Disease control rate DCR describes the percentage of patients with advanced cancer whose therapeutic intervention has led to a complete response, partial response, or stable disease [ 44 ].

DCR is related to ORR and has the greatest utility in evaluating cancer therapies that have predominating tumoristatic effects rather than tumoricidal effects [ 34 ]. For example, DCR has been found to predict subsequent survival in extensive stage small cell lung cancer in phase II clinical trials [ 45 ]. However, the use of DCR in phase II clinical trials offers little insight to the clinical benefit patients may receive from this therapy. Moreover, some argue that DCR might provide ambiguous information and potentially exaggerate the anticancer-effect of the therapy [ 44 ].

Clinical benefit rate CBR is defined as the percentage of advanced cancer patients who achieve complete response, partial response, or at least six months of stable disease as a result of therapy [ 44 ]. It has been argued that CBR should not be used at all since it does not measure clinical benefit [ 44 ].

There is also support to its use since like DCR, CBR can be useful for the rapid assessment of anticancer activity and can further capture disease stabilization [ 44 ]. Health-related quality of life HRQoL is an important measure that is patient reported and demonstrates clinical benefit. Quality of life is often used as a secondary clinical endpoint to compare treatments that have similar effects with differences in toxicity, but it can also be used as a co-primary endpoint with OS [ 4 , 47 ].

The four core questions include concepts related to overall health, physical health, mental health, and activities of daily living [ 48 ]. The items from this survey are brief and provide results that are easy for the assessor to interpret [ 48 ]. Milestone survival is defined as the survival probability at a given time point [ 9 ]. It is classified as an endpoint related to OS, however, it is considered a qualitative endpoint that describes a cross-sectional assessment at a specific time point and does not account for all OS data [ 49 ].

Milestone survival remains a possible surrogate endpoint for OS in late-stage drug development and, similar to the other endpoints used to evaluate cancer immunology advancements, requires further validation [ 49 ].

Table 1 summarizes the above described endpoints and Figure 1 depicts their relationship with tumor recurrence. Illustration of various endpoints with relationship to hypothetical tumor size on a time scale. The development of targeted therapies has been followed by a shift in the usage of primary clinical endpoints utilized in the evaluation of cancer drugs. The FDA initially used ORR to approve cancer drugs in the s, relying on imaging studies and physical assessments to measure outcomes [ 27 ].

OS proved to be a superior clinical endpoint for measuring direct clinical benefit in oncologic trials. Despite this finding, ORR remains a common surrogate clinical endpoint for cancer drugs under consideration for accelerated approval [ 50 ].

Special consideration for ORR as a primary clinical endpoint is also given in single-arm trials of patients with refractory tumors and no current therapy options [ 51 ].

This number decreased to Financial and time constraints continue to facilitate this shift since OS requires longer trials with larger numbers of patients compared to surrogate endpoints including PFS and DFS for metastatic and curative cancers, respectively. From to , DFS was used as a primary end point in five of the eight United States approved adjuvant or curative drugs in solid tumors [ 20 ].

Moreover, there is growing evidence for using PFS and DFS as primary clinical endpoints in special circumstances as outlined above. In a recent announcement by the FDA regulation 21CFR, subpart H , PFS and other surrogate endpoints can be used in cases to accelerate approval of drugs targeting serious or life-threatening diseases [ 53 ].

While prolonging survival is of primary goal of cancer treatments, the value of additional clinical endpoints to support OS is becoming clear. Part of chronic disease management includes mental health support, and clinical trials that can tell patients what to expect with regards to their disease state are of particular relevance. Additionally, with increasing interest in immuno-oncology, new endpoints will need to be explored starting from early-phase trials with MTD and MinED.

These, paired with milestone survival in late-phase trials, will allow researchers to evaluate the nonlinear dose-response and dose-toxicity kinetics characteristic of novel immunotherapy treatments [ 9 ]. Since immunotherapy is still a relatively new concept, validation of these unique endpoints will be critical to assessing future studies. Endpoints will continue to evolve not only with the development of new therapies but also as current imaging and detection modalities are defined.

As clinicians become better at identifying and classifying tumors, it is possible that patients who might have once been labeled with a stable disease would be found to have micrometastasis in the future. While current endpoints might capture future changes in imaging, some might start to fall out of favor while others increase in use to safely accelerate drug approval.

As more endpoints continue to be developed for specific types of cancer and specific therapies, it is important that studies clearly define which endpoints they are using and be able to differentiate them from other endpoints. Clinical Trial Endpoints. Here are a few common endpoints for clinical trials of cancer treatments: Disease Free Survival DFS — The length of time between treatment and relapse.

This endpoint is generally used with treatments that leave patients without any detectable signs of disease, such as surgery. Generally used to study advanced diseases that are unlikely to be removed entirely. Response Rate RR — The percentage of patients whose cancer shrinks or disappears after treatment.

Frequently used in single-arm trials , as it does not require direct comparison with a control. Overall Survival OS — The time between treatment and death. This measure includes death from any cause, including both the disease being treated and unrelated conditions.

Endpoints are measurable clinical and biological findings that are used for the development and assessment of treatment options. In the treatment of cancer, endpoints can be classified into two categories: "patient-centered clinical endpoints" including overall survival OS and health-related quality of life QoL , and "tumor-centered clinical endpoints" such as progression-free survival.

Surrogate endpoints are tumor-centered clinical endpoints that can be used as substitutes for patient-centered clinical endpoints, particularly OS. The choice of endpoints in oncology trials is a major problem.