Picture a site coordinator at a respiratory research center, mid-July 2026, three weeks out from her next monitoring visit. Her eTMF is current, her subject diaries are reconciled, and she has two patients scheduled for their Week 12 visits in CALM-2, GSK’s Phase 3 trial of camlipixant for refractory chronic cough. Then the email arrives: GSK is terminating the program. Development of camlipixant is ending. All sites will be contacted with closeout instructions. The coordinator has a PI who needs to notify patients, a pharmacy holding investigational product, a central IRB filing that needs an amendment, a contract with milestones that will never be reached, and no one on the sponsor line who can answer her questions right now.
This is the operational reality of a late-stage termination following a major acquisition. GSK announced in July 2026 that it is ending development of camlipixant after the CALM-1 and CALM-2 Phase 3 trials produced inconsistent results: CALM-1 met its primary endpoint at Week 12, CALM-2 did not reach statistical significance. GSK concluded that the limited efficacy demonstrated is unlikely to transform patient care in RCC. That judgment is reasonable. What follows it operationally, at hundreds of active sites, is anything but straightforward.
The Acquisition Premium Does Not Fund the Closeout
GSK completed the acquisition of Bellus Health on June 28, 2023, for US$2.0 billion, paying US$14.75 per share, a 103% premium to Bellus’ closing price on April 17, 2023. That premium bought camlipixant’s Phase 3 program, its P2X3 receptor antagonist mechanism, and an RCC market opportunity that analysts estimated could justify the price tag. What it did not buy, and what no acquisition model ever prices in, is the cost of an organized, sponsor-led closeout infrastructure if Phase 3 fails. That cost lands on the sites.
When a program terminates mid-execution, sites face a sequence of parallel tasks that their startup agreements almost never fully fund. IP accountability and destruction or return workflows begin immediately, governed by the investigational product accountability requirements in ICH E6(R3) Section 5.14. Central lab kits need reconciliation before the vendor contract closes. Source documentation that was built to support ongoing monitoring now needs to be organized to support a final close-out visit, a process that coordinators who were planning for long-term retention visits have not been staffing toward. The TMF reference model elements that were actively accumulating, training logs, deviation records, protocol amendment files, now need to be frozen and audited for completeness before eTMF access is revoked.
The IRB workflow alone creates a compressor on the coordinator’s calendar. A study termination requires a formal amendment or closure notification to the reviewing IRB, whether central or local. Sites using a central IRB for the CALM trials face a specific timing problem: central IRB submissions for termination are processed on a queue, not on the site’s preferred timeline, and the IRB’s acknowledgment letter is a required TMF document before the site file can be closed. At sites where that letter arrives 14 to 21 days after submission, the closeout visit cannot be completed, which delays final payment against the budget. For sites carrying 10 or more active subjects at termination, the timeline from “program terminated” email to final close-out visit completion routinely runs 60 to 90 days, during which the site is performing real work against a budget that stopped accruing milestones.
There is a counterintuitive pattern worth naming here. Sites that were performing well, high enrollment, low deviation rates, clean TMF, are often the hardest hit by a termination. They have the most subjects to transition off therapy, the most visit data to reconcile, and the most outstanding queries to resolve before database lock. The site that struggled to enroll three patients closes out in a weekend. The site that enrolled twenty absorbs weeks of uncompensated administrative burden. Budget structures that pay per-patient and per-visit milestones do not account for this asymmetry, and most sites discover it only when the termination letter arrives.
What the CRO Transition Actually Looks Like on the Ground
Acquisitions like the Bellus deal typically involve a CRO handoff at some point in the transition. Bellus ran early-stage operations; GSK’s internal clinical operations team or its preferred CRO absorbed the Phase 3 program. When the program terminates, the CRO’s resource model collapses quickly: CTMs are reassigned, CRAs lose their monitoring portfolios, and the site-facing contacts that coordinators built relationships with over 12 to 18 months of study execution disappear within weeks. Sites that call the CRA line post-termination to ask an operational question, “do I return the kits to the central lab or destroy on site?”, “who approves my final deviation report?”, are often routed through a generic inbox or told to await written instructions.
Across our network, the sites that navigate terminations most cleanly are the ones that do not wait for sponsor instructions before starting internal reconciliation. The coordinator who begins a self-audit of her investigational product accountability log on Day 1 of the termination notice, rather than Day 30 when the closeout visit is scheduled, is the one whose final visit takes three hours instead of eight. That is not a reflection of sponsor competence or incompetence. It is a reflection of the fact that closeout infrastructure is always underdeveloped relative to startup infrastructure, in every program, at every sponsor.
The CALM-2 failure also surfaces a specific problem for sites that enrolled under the assumption that a Phase 3 data package would support regulatory submission, and potentially a commercial launch, within a defined horizon. Refractory chronic cough affects an estimated 1.3% to 2.5% of the general adult population, and the Phase 2b SOOTHE trial had shown statistically significant reductions in cough frequency that justified advancing to Phase 3. PIs who positioned their sites as respiratory centers of excellence, who recruited coordinators on the promise of a growing RCC portfolio, now face a different conversation with their patients and their staff. The operational relationship between a promising Phase 2b signal and a Phase 3 termination is not a statistical anomaly. It is a recurring structural pattern in drug development that site contracts almost never acknowledge as a contingency.
What Sites and Sponsors Need to Do Monday Morning
For site teams receiving a termination notice right now: pull your IP accountability log, your subject status tracker, and your outstanding query list before you respond to any sponsor communication. Know your numbers before the first closeout call. Identify every subject who was on active treatment and map the notification timeline your IRB requires for early termination disclosures. Your central IRB submission for study closure needs to go in this week, not when the sponsor sends the template, because the acknowledgment letter is on the IRB’s queue, not yours, and it is on the critical path to your final payment.
For sponsors managing a termination of this scale: the coordinator on the ground does not need a letter of regret. She needs a close-out instruction manual within 72 hours of the termination announcement, a named CRA contact who will not be reassigned for 90 days, a clear answer on IP disposition, and a budget amendment that funds the actual closeout work rather than treating it as absorbed into existing milestones. Sites that are treated as operational partners during a termination become sites that answer your feasibility questionnaire when the next program starts. Sites that spend 90 days chasing a final payment invoice do not.
GSK has another 27 assets in late-stage development across its pipeline, and the reallocation of resources from camlipixant’s program will eventually surface somewhere in clinical operations as a new startup. The decision to terminate was framed around efficacy and patient impact, which is the right framing scientifically. The sites that absorb the operational fallout in the next 90 days will remember whether the operational execution matched that standard, and their feasibility responses on the next respiratory program will reflect it.
References
- Endpoints News — “GSK drops Phase 3 chronic cough drug in a blow to $2B Bellus Health buyout”
- GSK Press Release — “GSK completes acquisition of BELLUS Health” (June 28, 2023)
- GSK Press Release — “GSK provides an update on CALM-1 and CALM-2 Phase III trials”
- BioSpace — “GSK Culls Cough Program for Drug at the Heart of $2B Bellus Bet After Phase 3 Fail”
- PMC — Phase 2b SOOTHE trial: camlipixant efficacy data in refractory chronic cough
