The US Clinical Trial Recruiting Pipeline

Spotlight · Continuity check

Two calls, one month later: the tirzepatide surge cooled, the Keytruda slide held

Last issue we made two specific calls. The first was tirzepatide as mover of the month, surging from 27 actively-recruiting US trials in March to 35 in May. In June it added just one — 35 → 36. The surge has flattened. But the spread underneath it continued: distinct sponsors running tirzepatide studies rose from 30 to 32, and the number of indications attached to those trials widened from 66 to 72. So the drug is still diffusing into new disease areas even though net new trial starts paused for the month. The rest of the GLP-1 class stayed quiet too — semaglutide added two (23 → 25) and retatrutide held at three. Read: the March-to-May acceleration looks like a two-month pulse rather than the start of a sustained ramp. Whether July resumes it is exactly the kind of thing a monthly cadence is built to catch.

The second call was a “plateau watch” on Keytruda, down from 447 to 431 recruiting trials across the two-month window. June extends the slide: 431 → 425, another −6, with the number of indications attached falling from 833 to 820. At roughly six to eight trials a month, the decline is steady rather than accelerating — consistent with a maturing IV franchise whose investigator-led combination studies are reaching their enrollment caps. Around it, the PD-1/PD-L1 field tilted slightly negative this month: nivolumab −3, atezolizumab −4, cemiplimab −2, durvalumab flat. The exceptions were the China-origin checkpoint inhibitors, which kept gaining — toripalimab +3 (81 → 84) and tislelizumab +2 (88 → 90).

One reversal is worth flagging against last issue. Sacituzumab tirumotecan (sac-TMT), the Merck/Kelun TROP2 antibody-drug conjugate we noted growing +4 in the March-to-May window, gave back three trials this month (25 → 22) on an unchanged five sponsors — a reminder that at these trial counts a single month of ADC movement is well within the noise band. That is the broader takeaway for June: two calls tested, both broadly holding, and a board that otherwise moved no more than registry churn would predict. A quiet month is still a data point.