PD-0016 - Tumor budding in colorectal cancer with regards to mismatch repair status – prognostic value?

Date 27 June 2014
Event World GI 2014
Session Poster discussion session III – Endoscopy
Topics Colon Cancer
Rectal Cancer
Translational Research
Presenter M. Karlberg
Citation Annals of Oncology (2014) 25 (suppl_2): ii5-ii13. 10.1093/annonc/mdu164
Authors M. Karlberg1, P. Ragnhammar1, C. Lenander1, D. Edler2, M. Hallström1
  • 1Karolinska Institute Department of Oncology-Pathology, Stockholm/SE
  • 2Department of Molecular Medicine and Surgery, Karolinska University Hospital Solna, Stockholm/SE



Tumor budding is considered an adverse prognostic factor in colorectal cancer (CRC). A poorer prognosis is also observed in CRC patients with a proficient mismatch repair (pMMR) status. The aim of this study was to assess tumor budding in pMMR versus deficient mismatch repair (dMMR) in CRC.


Tumor budding was retrospectively examined using immunohistochemistry with pan-cytokeratin MNF-116 staining in the primary tumor from 134 patients with stage II and stage III CRC with known MMR status. The patients were included in an adjuvant Nordic trial between 1991 and 1996 randomly assigned to either surgery or surgery plus adjuvant 5-Fluorouracil (5-FU)-based chemotherapy. A total of 29 patients with tumors classified as dMMR status that had developed recurrence or distant metastases (dMMR/Met+) were matched according to age, gender, stage and treatment to 33 patients with dMMR status that had no recurrence or metastases (dMMR/Met-), 42 patients with pMMR status and no recurrence or metastases (pMMR/Met-) and 30 patients with pMMR and recurrence or distant metastases (pMMR/Met+). Assessment of tumor budding with MNF-116 staining was done by two independent investigators (MK, CL) counting at x20 magnification the number of tumor buds at the invasive tumor margin. Using quartile cut-offs the degree of budding was categorized into low-grade and high-grade tumor budding.


In the entire study population we found median budding to be 5 and the range 0-35 (SD ± 6). There was substantial agreement between the two investigators (MK and CL) on the presence of tumor budding p < 0.001, r = 1.0. In the lower quartile cut-off of 3 a tendency to high-grade tumor budding (≥3) was seen in the pMMR groups (Met+ and Met-) compared to the dMMR groups (Met+ and Met -) (p = 0.09) and in the subgroup pMMR/Met- compared to dMMR/Met- (p = 0.09). Using the median cut-off of 5 and upper quartile cut-off of 10 a higher percentage of high-grade tumor budding (≥5 and ≥10) was only found in the subgroup dMMR/Met+ compared to pMMR/Met+ (72% vs 40%, p = 0.01 and 45% vs 17%, p = 0.02, respectively). Regardless of cut-off used no survival advantage was observed in the entire cohort for the low-grade tumor budding group compared to the high-grade group.


In this study a significant higher number and grade of tumor budding was observed in the dMMR/Met+ group. This suggests that tumor budding could provide more accurate prognostic information for patients with a dMMR status and aid in the clinical decision of adjuvant chemotherapy especially in Stage II CRC patients.