This page presents the results of a rapid, living review of the clinical trial landscape for CTLA-4 immunotherapies. The data presented are derived from a search (conducted on September 22, 2019) of the U.S. National Library of Medicine’s clinical registry,, for all cancer trials testing a CTLA-4 immunotherapy. The registry search results were then combined with an automated search of the PubMed literature database to look for any publications reporting study results. Judgments about the outcomes of CTLA-4 trials (e.g., “positive” or “negative”) were based on the results and conclusions reported in these publications.

This analysis is intended to be a “living review”—in the sense that the data, presentation, and interpretations may change over time in order to reflect the most up-to-date and accurate information about this portfolio of research. The complete data set is available to browse at the bottom of the page. If you have questions, would like to contribute to the review, or discover errors in the data, please do not hesitate to reach out via e-mail (contact Spencer Hey,

The Basics

Total Trials






The numbers above give a high-level picture of the landscape. As of the last search date, there are 283 registered trials testing one of 15 different CTLA-4 targeting interventions. Most of the trials in this landscape (74%) are still open and actively recruiting patients. All told, these trials have enrolled (or are planning to enroll) 33,893 patients. To date, only 43 trials (15%) successfully completed. Fifteen trials were terminated (i.e., closed or cancelled before completing their target enrollment); 9 trials were withdrawn (i.e., terminated before enrolling any patients); and 2 trials were suspended (i.e., placed on hold, but not necessarily terminated). Searching the trial’s registration number identified 29 publications that reported trial results.

Exhibit 1: Clinical Trial Landscape of CTLA-4 Interventions

The figure below depicts all of the cancer clinical trials of CTLA-4 interventions, organized by the particular intervention (along the y-axis) and the trial’s start data (along the x-axis). The shape of the nodes in this figure corresponds to the trial’s status: circle = completed; circle with an “x” = terminated; triangle = active; square = not-yet-active, withdrawn, or suspended. Color indicates the outcome (if known): green = positive; red = negative/failed; yellow = unclear; gray = unknown; blue = ongoing. Size indicates number of patients enrolled. You can hover over any node to see more information (e.g., sponsor, malignancy, randomization, masking) and use the tools on the right side of the figure to zoom, pan, or download an image. You can also click on a node to open the trial’s corresponding registration page.

Bokeh Plot

What does this show?

Exhibit 1 shows that the vast majority of research activity in this landscape is focused on just 2 compounds: ipilumumab (186 trials) and tremelimumab (81 trials), whose research programs go back to 2001 and 2004, respectively. The other 13 compounds are all relatively new entrants into this space with only 1 trial that is still actively recruiting.

The figure also shows that the bulk of activity for both ipilimumab and tremelimumab is ongoing (i.e., the most dense areas consist of blue, triangular nodes in the past 5 years or so). Further, it is apparent that while ipilimumab has some large positive, randomized-controlled trials (RCT) supporting it, tremelimumab has yet to demonstrate any positive utility in an RCT.

Exhibit 2: Indication Landscape for Ipilimumab and Tremelimumab

The figure below provides a different perspective on the same ipilumumab and tremelimumab data. The trials are still arranged by start date along the x-axis, but the y-axis is now the type of malignancy. The rest of the coding scheme stays the same (i.e., shape = status; color = result; size = enrollment). However, this figure overlays trials from these two compounds, showing the how the research activity for these two compounds explored the space of possible cancer types. You can now click on the compound name in the legend at the bottom left to show/hide all the trials for that intervention.

Bokeh Plot

What does this show?

Exhibit 2 allows us to see how the research activity for these two interventions has explored the space of possible cancer indications. For example, we can see how the large positive trials of ipilimumab are all in melanoma (for which it received an FDA approval in 2011). But there has yet to be any similarly positive results in other cancer indications. The total portfolio of ipilimumab research is also remarkably consistent with other approved cancer drug portfolios—that is, early identification of efficacy in the indications that go on to gain regulatory approval followed by a long tail of exploration that does not (yet) demonstrate clinical utility.

Tremelimumab’s portfolio is particular striking in this respect, since it shows a similar pattern to ipilimumab, in terms of an early focus in melanoma later followed by a vigorous exploration of the indication space. However, in contrast with ipilimumab, there was no positive signal in the early phase 3 trials. This raises questions about the justification for all of the active tremelimumab trials. At least some of ipilimumab’s current exploratory activities could be justified by past signals of patient benefit (i.e., “We know that this treatment can work, so now we are trying to figure out the boundaries around its clinical utility.”) But since tremelimumab has yet to show any clear sign that it can be beneficial, what is a plausible justification for all of these active trials? The landscape suggests that it does not have similar efficacy to ipilimumab. So what exactly is the scientific, clinical, and ethical rationale for testing it across so many indications at once?

Exhibit 3: Outcome Landscape for Ipilimumab and Tremelimumab

The figure below provides yet another perspective on this data, sticking with the malignancy along the y-axis, but now arranging the trials by their primary outcome along the x-axis. This graph provides a snapshot of how much is known—and about what outcomes—for each intervention/malignancy combination. The coding scheme and the hovering function for the nodes is the same as above and clicking on the legend will again show/hide the trials for the respective compound.

Since some trials have multiple primary outcomes, this analysis places the node in the column of “highest clinical relevance” (i.e., If a trial used both overall survival and response as primary outcomes, the node would be placed in the overall survival (OS) column).

Bokeh Plot