Multicenter Validation Study of Pathologic Response and Tumor Thickness at the Tumor–Normal Liver Interface as Independent Predictors of Disease-Free Survival after Preoperative Chemotherapy and Surgery for Colorectal Liver Metastases
Atin Agarwal, Antoine Brouquet, Giuseppe Zimmitti, Scott Kopetz, Judith Stift, Catherine Julie, Anne-Isabelle Lemaistre, Stephane Benoist, Bernard Nordlinger, Alessandro Gandini, Michel Rivoire, Stefan Stremitzer, Thomas Gruenberger, Jean-Nicolas Vauthey, Dipen Maru. Baylor College of Medicine, Houston, TX; Ambroise Paré Hospital, Paris, France; University of Texas MD Anderson Cancer, Houston, TX; Medical University of Vienna, Vienna, Austria; Centre Léon-Bérard, Lyon, France
Background: To validate previously described pathologic markers of response to preoperative chemotherapy as predictors of disease-free survival (DFS) after resection of colorectal liver metastases (CLM).
Design: One hundred seventy one patients who underwent resection of CLM after preoperative chemotherapy at 4 major hepatobiliary centers were studied. Pathologic response defined as proportion of tumor cells remaining (categorized as complete;no residual tumor cells, major;1-49% residual tumor cells or minor;≥50% residual tumor cells) and tumor thickness at the tumor normal interface (TNI) (categorized <0.5mm, 0.5mm-<5 mm and ≥5mm) were assessed by blinded review by central pathology reviewer and pathologists from each institution.
Results: Oxaliplatin based regimen±bevasizumab was given in 62%,Irinotecan based regimen±bevasizumab in 28% and >1 regimens in 10% patients. Liver resection was minor in 46% and major in 54%. Pathologic response was complete in 8%, major in 49% and minor in 43%. Tumor thickness at TNI was <0.5mm in 21%, 0.5mm-<5mm in 56% and ≥5mm in 23%. Eight patients had positive margin. By log-rank analysis, survival difference between complete and major response(p=.004), major and minor response(p=.049), tumor thickness at TNI .5 mm and .5-<5 mm(p=.002), and tumor thickness at TNI .5-<5 mm and ≥5mm(p=.002) were significant. In univariate analysis, factors associated with worse 3 & 5years DFS were preoperative CEA (p=.005), minor vs. major or complete response, major vs. complete response, tumor thickness at TNI >0.5mm,>3 months of preoperative chemotherapy, and positive margin. In Cox regression multivariate analysis minor (p=.002) and major (p=.009) response, tumor thickness at TNI >0.5 mm (p<.001), >3 months of chemotherapy (p=.043),>3 tumor nodules (p=.037), and positive margin (p=.016) remained significant. Tumor size was predictor of pathologic response. Predictors of tumor thickness at TNI were tumor size, oxaliplatin based regimen and inclusion of bevasizumab. Agreement between central reviewer and local pathologists was near perfect(κ=.82) for pathologic response and substantial(κ=.76) for tumor thickness
Conclusions: Pathologic response and tumor thickness at TNI are valid predictors of DFS after preoperative chemotherapy and surgery for CLM.
Monday, March 4, 2013 11:30 AM
Proffered Papers: Section D, Monday Morning