Leveraging the perioperative window in pancreatic cancer: lessons from the PROSPER trial

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal solid malignancies. Despite the widespread use of adjuvant multi-agent chemotherapy, recurrence after curative-intent resection is near universal, with distant metastasis representing the predominant pattern of failure (1,2). In the PRODIGE-24 trial, modified FOLFIRINOX improved disease-free survival (DFS); however, most patients ultimately relapsed despite margin-negative resection (3). Similar findings were observed in ESPAC-4, underscoring that systemic therapy alone could not eliminate early metastatic recurrence (4). Collectively, these outcomes highlight a persistent biologic vulnerability in the postoperative setting where minimal residual disease likely determines long-term survival.

In parallel, surgery is not a biologically neutral event. The perioperative period is characterized by measurable physiologic perturbations, including a surge in catecholamines driven by adrenergic sympathetic activation, cyclooxygenase-mediated prostaglandin release in response to tissue injury, and transient suppression of cell-mediated immunity—particularly natural killer and cytotoxic T-cell function. A growing body of literature suggests that coordinated neuroendocrine and inflammatory responses may promote angiogenesis, enhance tumor cell invasiveness, facilitate epithelial-mesenchymal transition, and support micrometastatic outgrowth (5-7). Horowitz and colleagues proposed that surgery-induced stress responses may inadvertently accelerate metastatic processes during a critical perioperative window (5). Hiller and colleagues further emphasized that, although much of the evidence remains preclinical or observational, the biologic rationale for perioperative immunomodulation merits prospective evaluation (6).

Within this framework, combined β-adrenergic blockade and cyclooxygenase-2 (COX-2) inhibition has emerged as an attractive therapeutic strategy to blunt the aforementioned physiologic changes following surgery. Ricon and Ben-Eliyahu described how catecholamine and prostaglandin signaling act synergistically to suppress antitumor immunity and promote prometastatic pathways (7,8). Preclinical models demonstrate that dual inhibition attenuates surgery-induced metastatic progression (8). In this context, the PROSPER trial was designed to determine whether short-course perioperative propranolol (non-selective β-blocker) and etodolac (COX-2 inhibitor) could be safely and feasibly integrated into the perioperative window for pancreatic cancer surgery, representing one of the first randomized trials to evaluate perioperative stress modulation in patients with operable PDAC (9).

PROSPER was a single-center, randomized, double-blind, placebo-controlled Phase II trial conducted in patients undergoing pancreatoduodenectomy for presumed PDAC. Participants were randomized in a 1:1 design to receive perioperative propranolol and etodolac or matched placebo. The intervention began 10 days prior to surgery and continued for a total of 25 days, spanning the preoperative and early postoperative periods. The primary endpoints were safety and feasibility, defined as serious adverse events (SAEs) within 90 days and adherence to study medication, respectively. DFS, overall survival (OS), and recurrence patterns were evaluated as exploratory outcomes, alongside translational analyses of immune and inflammatory biomarkers.

The trial was terminated early due to slow accrual. Of 26 randomized patients, 20 ultimately received study medication, with nine assigned to active treatment and eleven to placebo. No SAEs attributable to the intervention were observed, and the overall rate of SAEs did not differ significantly between groups. Adherence to intervention was lower in the treatment arm, particularly in the postoperative period (treatment 50%±26% vs. placebo 78%±29%), although this difference did not reach statistical significance, likely due to inadequate power. While the study was not powered to assess oncologic efficacy, exploratory analyses suggested numerically longer DFS in the treatment group (16.4 vs. 11.3 months) and a markedly lower incidence of distant recurrence (11% vs. 55%), with OS also appearing favorable in the treatment arm (not reached vs. 15.8 months), although these findings must be interpreted with considerable caution given the small sample size and premature trial closure (9).

Translational analyses demonstrated attenuation of postoperative immune suppression and trends toward reduced inflammatory signaling among treated patients. Surgery produced the expected pattern of perioperative immune perturbation, characterized by expansion of myeloid populations and decline of adaptive immune effectors, including CD4⁺ and CD8⁺ T cells and natural killer cells. In patients receiving propranolol and etodolac, this decline appeared partially mitigated, with relative preservation of circulating T- and NK-cell populations in the early postoperative period. Consistent with this pattern, inflammatory cytokine levels—including IL-6—trended toward more rapid postoperative normalization in treated patients. Tumor immune analyses similarly suggested a directional increase in intratumoral immune cell density in treated patients. However, these findings did not reach statistical significance and must be interpreted cautiously given the small number of evaluable samples.

The PROSPER investigators should be commended for undertaking a prospective randomized trial leveraging the perioperative window in operable PDAC; in the authors’ opinion, this window is the optimal platform to advance both novel therapeutics and biomarker discovery in pancreatic cancer. Although the trial was terminated early and not positioned to demonstrate efficacy, its value lies in revealing constraints that must inform the design of the next generation of perioperative studies in PDAC. Despite screening more than 1,000 patients, only 26 were randomized and 20 initiated study medication. This discrepancy warrants closer scrutiny and represents a key takeaway for informing future trial design. While β-blocker use (~30%), drug interactions, and patient or provider concerns are cited as contributors, these factors alone do not reconcile the large gap between screened and randomized patients. Notably, the trial does not report how many screened patients were truly eligible, nor provide a breakdown of exclusions, limiting interpretation of where attrition occurred. The absence of these data represents an important limitation and constrains the ability to draw actionable lessons for future trial design. Although preexisting β-blocker use is cited as a major barrier to enrollment—a predictable challenge in an older population—it does not adequately account for the magnitude of patients not recruited. PROSPER is instructive not simply for what it tested, but also for what it exposed: that perioperative trials in PDAC must carefully balance biologic rigor with clinical pragmatism if they are to be feasible and scalable.

Adherence declined primarily during the postoperative period, likely reflecting the fragility of patients in the early postoperative period after pancreatoduodenectomy. Although delayed gastric emptying and the burden of taking multiple oral agents may have contributed, the identical dosing schedule in both arms suggests that medication logistics alone do not fully account for the differential adherence observed. Postoperative physiological vulnerability, including labile hemodynamics, fatigue, or β-blocker-related effects, may also have influenced treatment continuation. From a safety standpoint, PROSPER is reassuring. Nevertheless, in a small cohort with incomplete adherence, this safety signal should be interpreted as encouraging rather than definitive.

Interpretation of efficacy outcomes must remain cautious given the lack of powered study and reported differences should be considered exploratory. The observed separation in distant recurrence curves is hypothesis-generating but statistically inconclusive, particularly in a markedly underpowered cohort that included biologically heterogeneous final pathologies [with four of 18 resected patients (>20%) not having PDAC on final histology]. While the trial did not specifically restrict enrollment to PDAC and inclusion of other malignant pancreatic histologies reflects the realities of preoperative diagnosis, this approach becomes problematic in the context of such a limited sample size, where even modest heterogeneity can meaningfully influence exploratory signals. While inclusion of alternative malignant histologies reflects the realities of preoperative diagnosis in pancreatic surgery, the presence of non-malignant or premalignant conditions (e.g., IPMN with high-grade dysplasia and autoimmune pancreatitis) introduces additional biological heterogeneity that may further confound interpretation. With only 18 resected patients, apparent differences in infrequent surgical events such as pancreatic fistula, post-pancreatectomy hemorrhage, and 90-day mortality must be interpreted with similar caution.

The translational analyses provide important biologic context for interpreting PROSPER. In the placebo group, surgery was associated with the expected pattern of perioperative immune perturbation in a pattern consistent with potential metastatic vulnerability (5,6). Within this framework, the relative preservation of circulating T- and NK-cell populations observed in patients receiving propranolol and etodolac, along with downtrending IL-6 levels, is directionally consistent with the proposed mechanism of coordinated β-adrenergic and COX-2 blockade (8). Tumor immune analyses, although limited by small numbers and incomplete specimen availability, similarly suggested a trend toward intratumoral immunomodulation in treated patients. Taken together, these circulating, inflammatory, and tumor-based correlatives are consistent with the trial’s biologic premise and provide mechanistic plausibility, even if they remain hypothesis-generating rather than definitive.

While PROSPER is not practice-changing and does not support perioperative β-adrenergic and COX-2 inhibition as a standard component of pancreatic cancer surgery, its importance lies less in efficacy than in what it teaches the field about trial design. Future perioperative studies in PDAC will likely need to be shorter, simpler, broader in eligibility, and multicenter in execution if they are to be feasible and adequately powered. More specific design refinements are also warranted. Future studies could incorporate prospective screening logs to define where attrition occurs across eligibility, exclusion, and refusal, a key limitation of the current study. Given that prior β-blocker use represented a major barrier to enrollment, protocolized cardiology co-management to assess whether selected patients can be safely transitioned to nonselective β-blocker may meaningfully expand eligibility. Because adherence was largely preserved preoperatively but declines postoperatively, alternative strategies such as preoperative-only or preoperative-plus-inpatient intervention windows should be considered. In addition, dose-titration or stepped dosing approaches, as well as inpatient substitution strategies during periods of limited oral tolerance, may improve feasibility in the immediate postoperative setting. Thus, PROSPER should be viewed not as a failure of concept, but as a mandate for redesign—one that preserves the biologic promise of the perioperative window while adapting it to the practical realities of surgical care in PDAC.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Gland Surgery. The article has undergone external peer review.

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0166/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0166/coif). J.D. reports grant funding from Cantargia AB and consulting fees from Boston Scientific. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Groot VP, Gemenetzis G, Blair AB, et al. Implications of the Pattern of Disease Recurrence on Survival Following Pancreatectomy for Pancreatic Ductal Adenocarcinoma. Ann Surg Oncol 2018;25:2475-83. [Crossref] [PubMed]
2. Groot VP, Rezaee N, Wu W, et al. Patterns, Timing, and Predictors of Recurrence Following Pancreatectomy for Pancreatic Ductal Adenocarcinoma. Ann Surg 2018;267:936-45. [Crossref] [PubMed]
3. Conroy T, Hammel P, Hebbar M, et al. FOLFIRINOX or Gemcitabine as Adjuvant Therapy for Pancreatic Cancer. N Engl J Med 2018;379:2395-406. [Crossref] [PubMed]
4. Neoptolemos JP, Palmer DH, Ghaneh P, et al. Comparison of adjuvant gemcitabine and capecitabine with gemcitabine monotherapy in patients with resected pancreatic cancer (ESPAC-4): a multicentre, open-label, randomised, phase 3 trial. Lancet 2017;389:1011-24. [Crossref] [PubMed]
5. Horowitz M, Neeman E, Sharon E, et al. Exploiting the critical perioperative period to improve long-term cancer outcomes. Nat Rev Clin Oncol 2015;12:213-26. [Crossref] [PubMed]
6. Hiller JG, Perry NJ, Poulogiannis G, et al. Perioperative events influence cancer recurrence risk after surgery. Nat Rev Clin Oncol 2018;15:205-18. [Crossref] [PubMed]
7. Neeman E, Zmora O, Ben-Eliyahu S. A new approach to reducing postsurgical cancer recurrence: perioperative targeting of catecholamines and prostaglandins. Clin Cancer Res 2012;18:4895-902. [Crossref] [PubMed]
8. Ricon I, Hanalis-Miller T, Haldar R, et al. Perioperative biobehavioral interventions to prevent cancer recurrence through combined inhibition of β-adrenergic and cyclooxygenase 2 signaling. Cancer 2019;125:45-56. [Crossref] [PubMed]
9. Hüttner FJ, Klotz R, Giese NA, et al. Pancreatic resection with perioperative drug repurposing of propranolol and etodolac - the phase II randomized controlled PROSPER trial. Langenbecks Arch Surg 2025;410:168. [Crossref] [PubMed]