Innovative study design in clinical research

Clinical trials bridge the gap between preclinical studies and marketed products, with a primary goal of determining safety and efficacy.

Over the years, clinical trial protocols have significantly developed in term of design and execution. Innovation in clinical trials design has resulted in the increase of the adoption of innovative design types such as adaptive, umbrella, basket, and hybrid study designs.

Let’s have a look to a few of the available designs.

Adaptive study designs

Adaptive study designs allow real-time adjustments based on accumulating data, optimizing efficiency, and enhancing trial outcomes.

Some examples of adaptive study designs are:

  1. Adaptive randomization: this occurs when the probability of treatment assignment changes according to assigned treatment of patients already enrolled in the trial.
  2. Group Sequential Design: the trial is divided into sequential stages, with interim analyses conducted at predetermined intervals. Depending on the results of these analyses, the trial can be modified in terms of sample size, treatment arms, or even terminated earlier for efficacy or futility.
  3. Continuous Reassessment Model: this study design continually updates the dose allocation based on safety and efficacy data from previously enrolled participants. This adaptive approach aims to determine the maximum tolerated dose while minimizing the number of participants exposed to potentially harmful doses.
  4. Seamless Phase II/III Design: this design integrates the Phase II dose-finding stage with the Phase III confirmatory stage into a single trial. Favorable doses identified in Phase II are transitioned into Phase III, reducing timelines and resource requirements.
  5. Dose Ranging Design: those explore multiple dose levels to establish the dose-response relationship. Adaptive dose adjustments allow identifying the optimal dose for further investigation in subsequent phases.
  6. Drop-the-loser: this allows the dropping of inferior treatment groups or adding additional arms.
  7. Biomarker-adaptive Design: they identify the most suitable target subpopulations, based on clinical observations or known biomarkers, and evaluate the effectiveness of the treatment on that subpopulation.
  8. Adaptive treatment-switching: the investigator can switch a patient’s treatment from an initial assignment to an alternative treatment based on evidence of lack of safety or efficacy.

These adaptive designs offer several advantages over traditional fixed designs, including: efficiency, flexibility, and cost effectiveness. Both the FDA ( and EMA ( recognize the potential benefits of adaptive study designs in advancing drug development and improving patient outcomes.

Hybrid Retrospective-Prospective Trials

Hybrid retrospective-prospective studies are an approach in epidemiological research that combine elements of both retrospective and prospective study designs. In these studies, data collected prospectively and retrospective datasets are integrated in a unique study design.

A retrospective study looks back at data that has already been collected, often through medical records, historical data, or pre-existing databases (e.g., disease registries). This design is beneficial for studying rare diseases, conditions with long latency periods, or generating hypotheses. However, it can be limited by the quality and completeness of past data and the potential for recall bias. A prospective study, on the other hand, involves the collection of new data going forward from the time the study begins. Participants are followed over a period of time, which allows for the direct observation of outcomes as they occur. This approach provides high-quality data and reduces recall bias, but it can be time-consuming and expensive.

In a hybrid retrospective-prospective study, researchers may initially use retrospective data to identify a cohort or a population of interest and then plan a prospective study to investigate future outcomes. This design can take various forms, such as:

  • Retrospective Cohort with Prospective Follow-Up: A cohort is identified based on past records, such as patients diagnosed with a specific condition at some point in the past. Patients with those characteristics are then recruited for further research and followed prospectively to observe future outcomes, such as disease progression or response to treatments.
  • Exposure-Outcome Linkage: Historical data is used to establish an exposure, like occupational hazards or lifestyle factors, and the study prospectively monitors the cohort for outcomes related to these exposures, such as development of chronic diseases.
  • Case-Control with Prospective Components: Initially, a case-control study is conducted retrospectively to identify cases and controls. These groups are then followed prospectively to gather additional data or to track the progression of the condition under study.

Some advantages of hybrid studies are:

  • Comprehensive data collection: by combining past and future data, hybrid studies can provide a more complete picture of the disease or condition;
  • Efficiency: retrospective components can quickly identify cohorts and exposures, reducing the time needed for data collection
  • Cost-effectiveness: utilizing existing data reduces the costs associated with the prospective follow-up
  • Hypothesis generation and testing: the design allows for the generation of hypotheses through retrospective analysis, which can then be tested prospectively.

Umbrella trials

Umbrella trials are specifically designed to evaluate multiple targeted therapies within a single disease, typically cancer. This innovative trial design has emerged in response to the increasing understanding of the molecular heterogeneity within cancers, allowing for more personalized treatment strategies.

An umbrella trial focuses on a single type of cancer but tests the efficacy of different treatments targeting specific genetic or molecular abnormalities within that cancer. Instead of the traditional method of conducting separate trials for each treatment, umbrella trials streamline the process by grouping patients based on their tumor’s genetic profile and assigning them to corresponding targeted therapies within one overarching study.

Patients diagnosed with a specific type of cancer are enrolled into the study. Following, each patient undergoes extensive genetic and molecular profiling to identify specific mutations, biomarkers, or molecular characteristics of their tumor. Based on the genetic profiling results, patients are stratified into different subgroups. Each subgroup corresponds to a particular genetic mutation or molecular target. Patients within each subgroup are assigned to receive a targeted therapy specifically designed to interact with their tumor’s genetic or molecular features. The effectiveness of each treatment is continuously monitored. The trial design allows for adaptations, such as adding new targeted therapies or modifying existing treatment arms based on interim results.

Umbrella trials facilitate the delivery of personalized treatments tailored to the genetic makeup of an individual’s cancer, potentially improving outcomes. By testing multiple treatments within one trial, umbrella designs streamline the research process, reducing the time and resources needed compared to conducting separate trials. The design generates extensive data on various subtypes of a single cancer, enhancing the understanding of the disease and the effectiveness of different therapies. The trial can adapt to new scientific discoveries, incorporating new treatment arms or modifying existing ones based on emerging data.

Umbrella trials represent a paradigm shift in clinical research, offering a more efficient and personalized approach to evaluating multiple targeted therapies within a single disease framework. While they present certain logistical and financial challenges, the potential benefits in terms of accelerated drug development and improved patient outcomes make them a promising avenue in the era of precision medicine.

Basket trials

Differently from umbrella trials that focus on a single type of cancer, basket trials are designed to evaluate the effectiveness of a single drug or treatment across multiple types of cancers that share a common molecular or genetic alteration. Instead of traditional trials, which typically study one drug in one type of cancer, basket trials group different cancers by their shared molecular characteristics, not by their tissue of origin.

Patients with various types of cancers are enrolled into the study. Each patient undergoes genetic and molecular profiling to identify specific mutations or biomarkers that the drug targets. Patients are then categorized into “baskets” based on their tumors’ genetic mutations rather than the type of cancer. For example, a basket might include patients with lung, breast, and colon cancer, all of whom have a specific mutation in the BRAF gene. All patients within a basket receive the same targeted therapy designed to interact with their shared mutation. The response to the treatment is monitored across the different cancer types within each basket to determine the drug’s efficacy.

Advantages of such studies are:

  • Personalized Treatment: Basket trials allow for the testing of therapies tailored to specific genetic mutations, regardless of cancer type, promoting personalized medicine.
  • Efficiency: By grouping different cancer types with the same genetic alteration into one trial, basket trials streamline the research process, potentially accelerating the development of effective treatments.
  • Broad Applicability: These trials can identify new indications for existing drugs, expanding their use across multiple cancers.
  • Data Generation: Basket trials provide comprehensive data on the effectiveness of a drug across a range of cancers, enhancing the understanding of its broad therapeutic potential.

Basket trials represent a significant advancement in the field of clinical research, offering a more flexible and efficient way to evaluate targeted therapies across multiple cancer types. By focusing on genetic mutations rather than the tissue of origin, these trials align with the principles of precision medicine, aiming to provide more effective, personalized treatment options for patients. Despite their challenges, the potential benefits of basket trials in accelerating drug development and expanding the use of targeted therapies make them a promising approach in modern oncology research.