The Promise of Umbrella, Basket and N of 1 Clinical Trials in Cancer

More than 1.6 million people in the U.S. were expected to be diagnosed with cancer last year and more than half a million were expected to die from the disease, according to the American Association for Cancer Research and the Thomson Reuters Incidence & Prevalence Database. The World Health Organization’s projections for 2035 are even more discouraging: 2.4 million new cases in the U.S. and 24 million worldwide. In the fight against cancer, both public and private research has had a powerful impact. This research has spanned from bench to bedside and has been focused on the prevention, diagnosis, treatment and cure of any of the multitude of malignancies known to date. Historically, new treatments have been developed to treat a specific type of cancer. Through the years, however, investigators have observed that there is a wide range of variability in patient response. More recently, these outcome differences have been related to differences within individual subjects, giving rise to the promise of personalized medicine.

It is very common to find clinical trials that analyze some particular molecular, genetic or epigenetic trait on the participants and correlate that to the clinical response. Moreover, a growing number of experimental treatments are indeed being developed against a target molecule. For instance, Erbitux (cetuximab), a monoclonal antibody that targets epidermal growth factor receptor (EGFR), did not increase survival in EGFR-positive colorectal cancer patients with K-Ras mutations compared to best supportive care alone, according to results from one phase II and two phase III studies. According to this evidence, the FDA’s approval for cetuximab was limited to K-Ras wild-type, EGFR-expressing metastatic colorectal cancer. These and similar scenarios have lead to the outline of so-called “molecularly informed clinical trials” or “molecular trials.” These are more powerful and efficient by design.


It has been known for decades that cancers from the same anatomical area and histological appearance can be divided into different molecular subtypes, such as HER2-positive or HER2-negative breast cancers. An “umbrella” trial focuses on treating patients diagnosed with the same malignancy using different therapies. The anticancer agent will be selected according to factors such as the presence/absence of a given mutation or the levels of a particular biomarker. Researchers at The University of Texas M.D. Anderson Cancer Center selected more than 300 patients with chemorefractory stage IIIB-IV or advanced-incurable non-small cell lung cancer (NSCLC) to participate in the phase II randomized BATTLE-1 trial (NCT00409968) (see Figure 1). After the tumor molecular profile was determined (EGFR mutation/copy number, KRAS/BRAF mutation, VEGF/VEGFR2 expression, RXRs/Cyclin D1 expression and CCND1 copy number), patients were randomized to a particular treatment (erlotinib alone or combined with bexarotene, vandetanib or sorafenib). According to the published results, the BATTLE-1 study showed an “impressive benefit from sorafenib among mutant-KRAS patients” as well as the “feasibility of its novel design for advancing personalized treatment of NSCLC”.


Figure 1: Representation of BATTLE and NCI-MATCH trials. Molecularly profiled patients are assigned to a particular targeted-therapy (pills) according to the presence, or absence, of a specific genetic aberration or protein expression (stars).

A lot of progress has been made in the treatment of melanoma patients with BRAF mutations compared to non-BRAF-mutated melanoma. In the “Stand Up to Cancer and Melanoma Research Alliance Dream Team” (SUC-MRA) melanoma umbrella trial, researchers from Yale University and the Translational Genomics Research Institute in Phoenix will select non-BRAF-mutated melanoma patients and will explore the efficacy of several therapies using a preclinical-clinical approach. They will profile several tumor cell lines and then test their sensitivity to about 100 drugs. Next, they will recheck their results in animal models (xenografts of cell lines and primary tumors). And finally, they will conduct the clinical trial on the basis of the preclinical results. It is expected to achieve a 30% greater tumor response compared to standard-of-care therapy. According to investigators participating in the study, experiments to screen drugs on the melanoma cells are ongoing, the clinical protocol is already drafted, agreement with pharmaceutical companies has been obtained for around 30 drugs (10 of them are investigational), and the first patients are already enrolled.