Surveillance of Small Side-Branch Intraductal Papillary Mucinous Neoplasms: Is Long-Term Follow-Up Necessary?

Technical Summary: Surveillance of Small Side-Branch Intraductal Papillary Mucinous Neoplasms

Problem Statement
With the increasing utilization of cross-sectional abdominal imaging, the incidental detection of pancreatic cystic lesions has become common. Among these, side-branch intraductal papillary mucinous neoplasms (SB-IPMNs) are frequently identified. While main duct and mixed-type IPMNs carry a high risk of malignancy (up to 60%) and are often resected, SB-IPMNs are typically managed with long-term surveillance via MRI/MRCP, often for 5 to 10 years, based on revised Fukuoka criteria. Current guidelines recommend surveillance for SB-IPMNs <1.5 cm. However, the specific risk of malignant transformation in very small lesions, particularly those measuring <1 cm and those ≤5 mm, remains a subject of clinical debate. Previous studies, such as Wehrle et al., have suggested conversion rates exceeding 1% even in subcentimeter cysts, prompting recommendations for extended surveillance. This study addresses the uncertainty regarding the necessity of long-term follow-up for the smallest SB-IPMNs, specifically investigating whether lesions ≤5 mm warrant the same rigorous surveillance as larger cysts.

Methodology
The authors conducted a multicenter retrospective cohort study across eight hospitals and associated imaging centers, covering the period from July 1, 2006, to July 20, 2024.

• Data Extraction: Researchers utilized 18 keywords to extract MRI/MRCP reports referencing SB-IPMNs from electronic health records (EHR), initially identifying 9,821 patients aged 18–89.
• Inclusion/Exclusion Criteria: Patients with less than two years of follow-up were excluded. Further exclusions were applied to remove patients with prior pancreatic surgery, those with cysts determined to be pseudocysts, main duct IPMNs, mixed IPMNs, or those with pancreatic adenocarcinoma/masses on initial imaging. Patients whose cysts were resected immediately after the first imaging were also excluded.
• Cohort Definition: The final cohort consisted of 800 patients. They were stratified into two primary groups based on baseline cyst size: Group A (<1 cm, n=354) and Group B (≥1 cm, n=446). A secondary sub-analysis stratified patients into Group C (≤5 mm, n=120) and Group D (>5 mm, n=680).
• Variables: Data extracted included cyst size, growth rate, presence of suspicious features (e.g., mural nodules, wall thickening, main pancreatic duct dilation), and pathology reports to confirm invasive carcinoma (IC).
• Statistical Analysis: Comparisons between groups were performed using Fisher's exact tests and Kruskal–Wallis rank-sum tests.

Key Results

• Malignant Transformation Rates: The overall rate of biopsy-proven invasive carcinoma was 1.1% (9 of 800 patients).
  • **Size Comparison (<1 cm vs. ≥1 cm):** There was no statistically significant difference in carcinoma conversion rates between Group A (1.1%) and Group B (1.1%) (p > 0.99).
  • Sub-analysis (≤5 mm vs. >5 mm): In the smallest subgroup (≤5 mm), the conversion rate was 0.8% (1 of 120 patients), compared to 1.2% (8 of 680 patients) in the >5 mm group. This difference was not statistically significant (p > 0.99).
• Growth and Features:
  • Cysts <1 cm demonstrated a significantly higher mean relative growth rate (32.7%) compared to those ≥1 cm (9.4%), yet this growth did not correlate with an increased risk of malignancy.
  • In the ≤5 mm subgroup, the mean growth rate was 44.4% versus 15.3% in the >5 mm group.
  • Suspicious features were identified in 25.7% of the <1 cm group and 32.7% of the ≥1 cm group (p = 0.04).
• Follow-up Recommendations: Recommendations for endoscopic ultrasound (EUS) or surgical/GI consultation were significantly more frequent in the ≥1 cm group (24.2%) compared to the <1 cm group (5.1%).

Significance and Claims
The authors claim this is the first study to specifically evaluate the malignant potential of SB-IPMNs measuring ≤5 mm. The primary significance of the findings lies in the observation that the conversion rate for lesions ≤5 mm (0.8%) falls below the 1% threshold commonly used in pulmonary nodule surveillance guidelines.

The paper argues that while current guidelines (revised Fukuoka) support long-term surveillance for subcentimeter cysts, the data suggests that size alone may be a less meaningful discriminator of malignant potential among SB-IPMNs, particularly for the smallest lesions. The authors note that the discordance between high relative growth rates in small cysts and low malignancy rates suggests that interval growth in these lesions may reflect earlier detection rather than biological aggressiveness.

Conclusion
The study concludes that while surveillance for SB-IPMNs >5 mm should continue according to current guidelines, there is a potential opportunity to reduce or de-escalate imaging follow-up for the smallest SB-IPMNs (≤5 mm). The authors emphasize that these findings could help reduce unnecessary healthcare utilization, costs, and patient anxiety associated with long-term surveillance of very small, incidental pancreatic cysts. However, they maintain a modest tone, acknowledging limitations such as the retrospective nature of the study, potential selection bias, the low number of carcinoma events limiting statistical power, and the lack of generalizability due to the predominantly White study population. They call for additional studies to further characterize these very small lesions and refine surveillance recommendations.