Artikel
CpG island methylator phenotype infers a poor prognosis in locally advanced rectal cancer
Suche in Medline nach
Autoren
-
Peter Jo
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
-
Klaus Jung
- Universitätsmedizin Göttingen, Medizinische Statistik, Göttingen
-
Marian Grade
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
-
Lena-Cristin Conradi
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
-
Hendrik Andreas Wolff
- Universitätsmedizin Göttingen, Strahlentherapie, Göttingen
-
Heinz Becker
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
-
Josef Rüschoff
- Pathologie Nordhessen, Pathologie, Kassel
-
Arndt Hartmann
- Universitätsklinikum Erlangen, Pathologisches Institut, Erlangen
-
Tim Beissbarth
- Universitätsmedizin Göttingen, Medizinische Statistik, Göttingen
-
Annegret Müller-Dornieden
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
-
Michael Ghadimi
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
-
Regine Schneider-Stock
- Universitätsklinikum Erlangen, Pathologisches Institut, Erlangen
-
Jochen Gaedcke
- Universitätsmedizin Göttingen, Allgemein- und Viszeralchirurgie, Göttingen
| Veröffentlicht: | 20. Mai 2011 |
|---|
Gliederung
Text
Introduction: Locally advanced rectal cancers are treated with preoperative radiochemotherapy (RT/CT). However, subsets of patients have no benefit from neoadjuvant treatment. Since epigenetic modifications, including DNA methylation, may influence response to neoadjuvant treatment we studied the CpG island methylator phenotype (CIMP) in patients who received a 5-fluouracil based RT/CT.
Materials and methods: Eighty-six patients with locally advanced rectal cancer, treated within a phase III clinical trial (CAO/ARO/AIO-94 and -04), were included in this analysis. CIMP was assessed by methylation specific PCR (MSP) using RUNX3, SOCS1, NEUROG1, IGF2 and CACNA1G as marker panel. Loss of mismatch repair gene (MMR) expression was assessed by immunohistochemistry and KRAS and BRAF mutation status was available from previous studies.
Results: The CIMP phenotype could be established in 78 out of 86 patients (90%). Five patients (6.4%) revealed CIMP positivity (≥ 3 methylated promoters), whereas 73 patients (93.6%) where classified as CIMP negative. None of the samples showed a loss of MMR expression, and a single mutation of the BRAF gene (D594G) was detected only once. Mutations within the KRAS gene (exon 1,2, and 3) were present in 48% of cases (n=35) and were not correlated to a specific CIMP status. However, CIMP positivity was associated with low tumor regression (p=0.02) following preoperative RT/CT. For both groups, disease-free survival was comparable (p=0.55), whereas overall survival was notably worse in CIMP positive patients (p=0.02) suggesting an increased likelihood of poor clinical outcome (Hazard Ratio 7.0; 95%-CI: (1.4, 36.5)).
Conclusion: CIMP positivity is an infrequent event in locally advanced rectal cancer. It increases the likelihood of a poor prognosis dramatically and is associated with poor response to preoperative RT/CT. Once validated in an independent larger patient cohort, the CIMP status may be included as a molecular marker for individualized treatment stratification.
