---

FLOT + Durvalumab vs. FLOT + Nivolumab in Gastric/GEJ Adenocarcinoma

Bottom Line Up Front

---

1. FLOT + Durvalumab: MATTERHORN Trial (Phase III)

• EFS benefit was statistically significant (primary endpoint met)
• OS benefit confirmed at final analysis (ESMO 2025), median OS not reached in either arm
• 6.7% absolute improvement in 36-month survival (68.6% vs. 61.9%)
• pCR rate nearly tripled (19.2% vs. 7.2%)
• Safety profile comparable to FLOT alone - no new signals
• FDA approved for perioperative resectable GC/GEJC based on these results

---

2. FLOT + Nivolumab: Phase II Data (Advanced/Metastatic Setting)

• FLOT + nivolumab showed "promising efficacy" and was feasible
• This was a phase II signal-finding study, not powered for definitive conclusions
• The trial also tested dual checkpoint blockade (nivolumab + ipilimumab) with mFOLFOX, which increased toxicity without improving efficacy
• No OS data reported from this arm
• Not FDA approved; no phase III data exist for this combination

---

3. Meta-Analysis Context (Phase III Trials)

• No significant difference in EFS between FLOT + durvalumab and FLOT + pembrolizumab (HR 0.907, P=0.586) - these two PD-L1/PD-1 combinations appear similar when added to FLOT
• FLOT + IO (any checkpoint inhibitor) was superior to cisplatin/fluoropyrimidine + IO (HR 0.790)
• Concluded: "Addition of IO to perioperative FLOT should be considered the preferred standard of care in resectable, locally-advanced gastroesophageal adenocarcinoma"

---

4. Why the Comparison Is Unequal

---

5. Clinical Takeaway

• For resectable gastric/GEJ adenocarcinoma: FLOT + durvalumab is the new standard of care based on MATTERHORN. It is FDA-approved, with proven EFS and OS benefit.
• For advanced/metastatic disease: nivolumab-based regimens (e.g., nivolumab + chemotherapy per CheckMate-649) are established, but the specific FLOT + nivolumab combination in this setting remains investigational. Standard first-line options for metastatic gastric cancer currently include nivolumab + FOLFOX/XELOX (particularly in PD-L1 CPS ≥5 tumors).
• There is no head-to-head trial comparing FLOT + durvalumab vs. FLOT + nivolumab. Based on the meta-analysis, FLOT-IO combinations likely perform similarly when the same chemotherapy backbone is used, but durvalumab has all the phase III evidence in the perioperative setting.

---

1. Janjigian YY et al. NEJM 2025 - MATTERHORN trial [PMID: 40454643]
2. Lorenzen S et al. Nature Commun 2026 - FLOT+nivolumab phase II [PMID: 41760643]
3. Zhao JJ et al. Cancer Treat Rev 2025 - Meta-analysis [PMID: 40779895]

---

PICC Line vs. Chemoport (Implantable Port): A Clinical Comparison

---

What Are They?

---

Complications: Evidence Summary

1. Thrombosis (Strongest Evidence)

• Port VTE risk is significantly lower than PICC (OR 0.38, 95% CI 0.25-0.58)
• The PICC's longer catheter in a smaller vein creates more stasis and endothelial injury - the primary driver of thrombogenesis

• Overall adverse effects: PICC worse (OR 2.72 favoring port)
• Catheter-related thrombosis: PICC worse (OR 2.84, P<0.00001)
• DVT and infection: no significant difference between groups

• Total complications: PICC significantly worse (RR favoring port, P=0.03)
• Deep vein thrombosis: PICC worse (RR 0.31, P=0.0009)
• Unplanned catheter removal: PICC worse (RR 0.57, P=0.0005)
• Local reactions/edema: PICC worse
• Port was worse for: pocket/exit-site infection without septicemia (RR 2.34) and pain at insertion site (RR 4.06)

2. Infection

• Overall catheter-related bloodstream infection (CRBSI): rates are similar between the two devices in most meta-analyses
• Ports have higher rates of pocket infection (subcutaneous space around reservoir)
• PICCs have higher rates of local/exit-site infection and skin reactions
• Sepsis rates are comparable

3. Summary Table of Complication Rates

---

Practical Considerations

---

Quality of Life

• Breast cancer meta-analysis (Chen et al., 2025, PMID 41163256, n=1,125): IPC associated with fewer complications (RR 1.75 favoring port, 95% CI 1.24-2.46), driven by higher thrombosis and infection rates with PICC
• Patients with ports report better body image, greater freedom with daily activities (bathing, swimming), and less anxiety about arm use
• PICC maintenance burden (frequent dressing changes) is a significant quality-of-life concern for ambulatory patients on long regimens

---

When to Choose Which

• Short chemotherapy course (4-12 weeks)
• Patient is not a surgical candidate or has bleeding disorders
• No suitable chest vein anatomy for port
• Rapid initiation needed (e.g., starting chemo within 24-48 hours)
• Patient preference for avoiding surgical procedure
• Pediatric or very thin patients where port reservoir may erode skin

• Long or multi-cycle chemotherapy (>6 months) - e.g., FLOT, FOLFOX, adjuvant regimens
• High thrombosis risk (obesity, prior DVT, hypercoagulable state)
• Patient values minimal lifestyle disruption and hidden device
• Vesicant chemotherapy agents requiring reliable central access
• Planned long-term surveillance with intermittent access needs
• Better quality of life priority

---

Bottom Line

• Lower overall complication rates
• Much lower thrombosis risk (OR 0.38 for port vs PICC)
• Better quality of life and fewer activity restrictions
• Lower long-term maintenance burden

---

1. Wang P et al. Thromb Res 2022 - VTE meta-analysis [PMID: 35279504]
2. Lin L et al. Front Oncol 2023 - PORT vs PICC meta-analysis, 22 studies [PMID: 37519803]
3. Chen W et al. Med Sci Monit 2025 - Breast cancer meta-analysis [PMID: 41163256]
4. J Cancer 2025 - RCT-based meta-analysis, n=1,127 patients

---

MATTERHORN Trial - Subgroup Forest Plot Analysis (Figure 1, Page 9/14)

---

PD-L1 Expression (TAP Score) - The Key Focus

• The TAP ≥1% group (n=853, the majority) shows a clear, statistically significant benefit from durvalumab: HR 0.70, with CI entirely or nearly below 1.0
• The TAP <1% group (n=95, small subgroup) shows a non-significant trend (HR 0.77), but the wide CI (0.40-1.46) crosses 1.0 - this reflects the small sample size, not necessarily absence of benefit
• Crucially, the OS data (ESMO 2025 final analysis) confirmed benefit regardless of PD-L1 status (TAP ≥1%: OS HR 0.79; TAP <1%: OS HR 0.79, both similar)
• Key message: PD-L1 expression (TAP score) does NOT reliably predict benefit from durvalumab in this perioperative setting. Treatment benefit is seen across the entire population.

---

Full Subgroup Analysis from the Image

Sex

• Women show a weaker trend than men; experts caution this may reflect smaller numbers and the diffuse-type predominance in women

Age at Randomization

• Benefit is consistent across age groups - younger and older patients both benefit equally (HR nearly identical: 0.71 vs. 0.70)

Geographic Region

• Non-Asian patients show clear significant benefit; Asian subgroup trend is consistent but CI crosses 1.0 due to smaller numbers (n=180)
• OS subgroup at ESMO 2025 confirmed benefit in both regions

Clinical Lymph Node Status

• Node-positive patients benefit more (HR 0.67, clear significance)
• Node-negative patients: benefit less clear (HR 0.85, CI crosses 1.0) - likely because they already have better prognosis with FLOT alone

Primary Tumor Location

• GEJ tumors show greater absolute and relative benefit (HR 0.61) vs. gastric (HR 0.76)
• Both locations benefit significantly

ECOG Performance Status

• Both PS 0 and PS 1 benefit; PS 0 reaches significance, PS 1 borderline (smaller n)

Histologic Type

• Intestinal-type benefits significantly (HR 0.66)
• Diffuse-type shows NO meaningful benefit (HR 0.93, CI spans 0.66-1.32) - this is a notable finding; diffuse gastric cancer (signet ring cell, linitis plastica) may be biologically less responsive to PD-L1 blockade
• Indeterminate type shows strong benefit (HR 0.56)

Microsatellite Instability (MSI) Status

• MSI-H subgroup is too small to draw conclusions (n=49) - NC (not calculable) reported
• MSS/not-high patients clearly benefit (HR 0.67) - this is the important finding because MSS tumors are the vast majority (~95%) of gastric cancers
• Interestingly, the OS data also confirmed MSS patients benefit, which is striking since MSS tumors are typically less immunogenic

---

Overall Summary Table

---

Clinical Takeaways

1. PD-L1 is NOT a required biomarker for durvalumab use in the perioperative setting - benefit is seen regardless of TAP score. This is a key difference from the metastatic setting.
2. Diffuse-type histology is the one subgroup where benefit appears absent (HR 0.93). This biologic signal deserves attention and may reflect the immunosuppressive tumor microenvironment of diffuse/signet ring cell cancers.
3. Node-negative patients benefit less, likely because their baseline prognosis with FLOT alone is already good.
4. GEJ tumors and intestinal-type derive the greatest benefit.
5. The consistency of benefit across most subgroups (even MSS, even older age, both geographic regions) is what solidifies durvalumab + FLOT as the new standard of care for resectable G/GEJ adenocarcinoma.

---

Can Durvalumab Be Given When PD-L1 TAP <1%?

Short Answer: YES

---

Why the TAP <1% Data Should Not Stop You

1. The Numbers Are Too Small to Draw Conclusions

• TAP <1% subgroup: only n=95 patients total (48 durvalumab, 47 placebo)
• This is ~10% of the total trial population (948 patients)
• With such a small n, the study is massively underpowered to detect a real difference
• Wide CI of 0.40-1.46 means the true HR could be anywhere from a strong 60% risk reduction to a 46% risk increase - statistically uninformative

2. The EFS Trend Still Favors Durvalumab

• TAP <1%: HR 0.77 - the point estimate still favors durvalumab
• TAP ≥1%: HR 0.70 - very similar point estimate
• The two HRs are nearly identical (0.77 vs 0.70), suggesting no meaningful biological difference by PD-L1 cut-off

3. OS Data Confirms Benefit Regardless of PD-L1

• TAP ≥1%: OS HR 0.79
• TAP <1%: OS HR 0.79
• Literally identical OS benefit in both PD-L1 subgroups

4. Perioperative Setting Behaves Differently from Metastatic

5. Regulatory and Expert Consensus

• The FDA label does not include a PD-L1 cut-off as a requirement
• ESMO 2025 discussants explicitly stated: "The treatment effect is independent of PD-L1 expression"
• Expert opinion (Klempner et al., ASCO 2025): "Durvalumab + FLOT is the right therapy for everyone"

---

Practical Algorithm

Resectable Gastric/GEJ Adenocarcinoma, Stage II-IVA
                    ↓
         Check PD-L1 (TAP score)?
                    ↓
    TAP ≥1%              TAP <1%
       ↓                    ↓
  Durvalumab + FLOT    Durvalumab + FLOT
  (clear benefit)      (same benefit per OS data)
                    ↓
       PD-L1 does NOT change the decision

---

The One Caveat

• Diffuse histology (HR 0.93 - no apparent EFS benefit)
• Node-negative disease (HR 0.85 - less clear benefit)

---

Bottom Line