Highlights
– PASS-01 randomized 160 patients with de novo metastatic PDAC to modified FOLFIRINOX (mFFX) versus gemcitabine/nab‑paclitaxel (GnP). Progression‑free survival (PFS) was similar; overall survival (OS) and safety trends favored GnP.
– Comprehensive molecular characterization (whole genome, transcriptome, and patient‑derived organoids) was feasible in most patients, but RNA subtype (basal‑like vs classical) provided only suggestive, nondefinitive predictive signals.
– Second‑line, correlate‑guided therapy was limited by short survival and brief treatment duration, arguing that precision strategies may need to be applied earlier in treatment.
Background: unmet needs in metastatic PDAC
Pancreatic ductal adenocarcinoma (PDAC) remains one of the deadliest solid tumors; most patients present with unresectable or metastatic disease and median survival measured in months. Two cytotoxic regimens—FOLFIRINOX and gemcitabine plus nab‑paclitaxel—improved outcomes compared with gemcitabine alone and are widely used first‑line options for fit patients. However, there is no universally accepted standard of care that is superior across all patients, and real‑world tolerability and outcomes vary. Molecular subtyping of PDAC (classical versus basal‑like) and tumor genomic profiling have raised hopes that predictive biomarkers could guide treatment selection, but prospective randomized data incorporating such correlatives are limited.
Study design: PASS‑01 at a glance
PASS‑01 (Knox et al., J Clin Oncol 2025) is a randomized phase II trial that enrolled 160 patients with previously untreated, de novo metastatic PDAC. Patients with germline pathogenic variants in BRCA1/2 or PALB2 were excluded to avoid confounding by known platinum/ PARP sensitivity. Participants were randomized 1:1 to receive modified FOLFIRINOX (mFFX) or gemcitabine/nab‑paclitaxel (GnP). The primary end point was progression‑free survival (PFS) between arms, evaluated with an unconventional alpha threshold (significance level 0.3) appropriate to this randomized phase II design intended to detect signals rather than provide definitive phase III evidence. The per‑protocol population included patients who received at least one dose of study chemotherapy.
Pretreatment biopsies underwent whole‑genome and transcriptome sequencing and attempts were made to establish patient‑derived organoids (PDOs). Molecular tumor board recommendations were provided, and outcomes were analyzed by RNA subtype (basal‑like versus classical) among other correlatives. Second‑line therapies were recorded, and use of correlate‑guided therapy was evaluated.
Key findings
Enrollment and follow‑up: Of 160 randomized patients (80 mFFX; 80 GnP), 140 comprised the per‑protocol population (71 mFFX; 69 GnP). Median follow‑up was 8.3 months.
Primary and survival outcomes
In the intention‑to‑treat analysis the median PFS was 4.0 months for mFFX versus 5.3 months for GnP (hazard ratio [HR] 1.37; 95% CI, 0.97–1.92; P = 0.069). Median overall survival (OS) was 8.5 months with mFFX versus 9.7 months with GnP (HR 1.57; 95% CI, 1.08–2.28; P = 0.017). Thus, while PFS differences were nonsignificant by conventional thresholds, OS was notably worse with mFFX in this cohort, with the authors reporting safety and tolerability trends favoring GnP.
Molecular correlatives and transcriptome subtypes
Feasibility of correlative assays was high: genomic data were obtained in 94% of cases, transcriptomes in 74%, and PDOs were successfully established for 50% of patients. Analysis by RNA subtype produced the following median PFS and OS results:
- Basal‑like: median PFS 3.0 months (mFFX) versus 5.5 months (GnP) (P = 0.17); median OS 7.5 months (mFFX) versus 8.9 months (GnP) (P = 0.75).
- Classical: median PFS 6.3 months (mFFX) versus 5.4 months (GnP) (P = 0.36); median OS 9.7 months (mFFX) versus 13.9 months (GnP) (P = 0.047).
These subgroup findings suggest a possible interaction whereby classical PDAC may derive greater OS benefit from GnP in this dataset, while basal‑like tumors had uniformly poor PFS. However, subgroup sizes were modest and P values borderline; the authors appropriately interpret these as hypothesis generating rather than definitive evidence of predictive biomarker utility.
Second‑line therapy and correlate‑guided treatment
Overall, 75 patients (54%) received second‑line therapy; 33 of these (44%) received correlate‑guided recommendations from the molecular tumor board. Median time on second‑line therapy was brief (2.1 months) and median OS from initiation of second‑line was 5.4 months with correlate‑guided therapy versus 4.4 months with standard chemotherapy (P = 0.45). The small and nonrandomized nature of this comparison, together with the short duration of exposure, argues that second‑line settings in unselected metastatic PDAC often provide too little time to derive benefit from precision approaches.
Safety
The published summary reports that safety trends favored GnP, consistent with the study conclusion. Specific grade‑level toxicity breakdowns and adverse event rates are reported in the original article; no fabricated detail is provided here. The differential tolerability likely contributed to survival differences, particularly in a population with high symptom and comorbidity burden.
Expert commentary: interpretation, strengths, and limitations
PASS‑01 brings a valuable randomized, biomarker‑rich perspective to a question clinicians face frequently: which multiagent cytotoxic regimen to choose first in metastatic PDAC. Strengths include randomized allocation, exclusion of germline BRCA/PALB2 carriers to avoid confounding by known platinum sensitivity, and extensive molecular and organoid correlative work demonstrating feasibility in a multicenter setting.
Limitations are important to recognize. This is a phase II trial not powered for definitive practice‑changing conclusions. The primary alpha of 0.3 reflects a signal‑seeking design; conventional thresholds were exceeded for OS but not for PFS. The per‑protocol population excluded patients who never started therapy, which can bias estimates if early dropouts differ between arms. The molecular subgroup analyses are underpowered and susceptible to multiple testing; the apparent OS advantage for GnP in classical PDAC (P = 0.047) is hypothesis generating and requires prospective validation.
Biologically, classical and basal‑like PDAC subtypes reflect distinct transcriptional programs and tumor microenvironment features; prior work (Collisson et al., Moffitt et al., Bailey et al.) has suggested prognostic and potentially predictive roles. PASS‑01’s data hint that transcriptional subtype could influence chemosensitivity, but the signal is modest and inconsistent across endpoints.
Importantly, the organoid and genomic pipelines worked well in the trial, but the clinical utility of post‑progression correlate‑guided therapy was limited by short survival and time on treatment, highlighting the practical need to bring precision strategies earlier (first line) or to improve rapidity of actionable output.
Clinical implications and next steps
For practicing clinicians, PASS‑01 suggests that GnP remains an appropriate—and potentially preferable—first‑line option for many patients with de novo metastatic PDAC, particularly when considering tolerability and the OS signal observed in this randomized phase II cohort. That said, regimen selection must be individualized for fitness, comorbidity, goals of care, and patient preferences.
Research priorities emerging from PASS‑01 include: conducting larger randomized trials stratified by validated RNA subtypes to test predictive hypotheses; exploring strategies to accelerate molecular/functional readouts so precision recommendations can be applied in the first‑line setting; and developing pragmatic approaches to integrate organoid chemosensitivity testing into decision making without delaying treatment initiation.
Conclusion
PASS‑01 demonstrates that in an unselected, BRCA/PALB2‑negative metastatic PDAC population, PFS did not differ meaningfully between mFFX and GnP, but OS and safety trends favored GnP. Molecular profiling and organoid development are feasible at scale, but current second‑line, correlate‑guided strategies offered limited clinical benefit in this setting because of short patient survival and brief treatment exposure. These results encourage prospective trials that incorporate molecular subtyping into first‑line randomization and faster translational pipelines to make precision oncology actionable earlier in the disease course.
Funding and clinicaltrials.gov
Funding sources and trial registration details are reported in the primary publication (Knox et al., J Clin Oncol 2025). Readers should consult the original article for specific sponsor and registration information.
References
1. Knox JJ, O’Kane G, King D, et al. PASS‑01: Randomized Phase II Trial of Modified FOLFIRINOX Versus Gemcitabine/Nab‑Paclitaxel and Molecular Correlatives for Previously Untreated Metastatic Pancreatic Cancer. J Clin Oncol. 2025 Nov;43(31):3355‑3368. doi:10.1200/JCO‑25‑00436.
2. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011 May 12;364(19):1817‑1825.
3. Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab‑paclitaxel plus gemcitabine. N Engl J Med. 2013 Oct 31;369(18):1691‑1703.
4. Collisson EA, Sadanandam A, Olson P, et al. Subtypes of pancreatic ductal adenocarcinoma and their differing responses to therapy. Nat Med. 2011 Apr;17(4):500‑503.
5. Moffitt RA, Marayati R, Flate EL, et al. Virtual microdissection identifies distinct tumor‑ and stroma‑specific subtypes of pancreatic ductal adenocarcinoma. Nat Genet. 2015 Oct;47(10):1168‑1178.
6. Bailey P, Chang DK, Nones K, et al. Genomic analyses identify molecular subtypes of pancreatic cancer. Nature. 2016 Mar 3;531(7592):47‑52.