Recently, evidence has accumulated that weekly 24-h infusion of high-dose 5-fluorouracil (5-FU) with leucovorin (LV, folinic acid) biochemical modulation may improve the response rates compared with the bolus 5-FU regimens in colorectal cancer (CRC). Combining the infusional 5-FU/LV (iFL) regimens with oxaliplatin or irinotecan is widely adopted to further improve treatment efficacy. Either oxaliplatin-iFL or irinotecan-iFL may achieve an overall response rate of more than 50% in the first-line treatment. Intriguingly, in the salvage treatment for metastatic CRC patients who had failed iFL, only oxaliplatin-iFL may achieve a response rate of about 13-25%. In contrast, oxaliplatin alone or irinotecan-iFL had a very low response rate of 5% or less. To test if the oxaliplatin may reverse the iFL-related 5-FU resistance in CRC, we used DLD-1 colon adenocarcinoma cells as the in vitro study model. First, we revealed that oxaliplatin and 5-FU act synergistically on DLD-1 cells by MTT cytotoxicity assay and median drug effect analysis. Second, we treated the DLD-1 cells with serial concentrations of oxaliplatin (0.1-10 microM). Oxaliplatin treatment results in down-regulation of free thymidylate synthase (TS) protein expression by Western blotting. Further, we analyzed the TS mRNA level by reverse transcription and real-time quantitative polymerase chain reaction assay. Oxaliplatin treatment results in down-regulation of the TS mRNA level up to 40% (mean +/- SD of ratio to reference control = 0.60 +/- 0.21, range 0.42-0.84). In this study, our data provide important information explaining the reason why the combination of oxaliplatin and 5-FU results in a better objective response in 5-FU-resistant patients than oxaliplatin alone does. Our data also suggest that TS down-regulation happens at the transcriptional level. TS modulation and down-regulation had, thus, shed light on the useful potential strategy to achieve objective responses in 5-FU-resistant CRC patients.

A 61-year-old never-smoker female suffered from adenocarcinoma of the lung with chest wall invasion and peri-adrenal lymph node metastases. After palliative resection of all clinically detectable primary and metastatic adenocarcinoma, she received cisplatin and gemcitabine combination chemotherapy for a total of 3 cycles. New metastatic lesions were found in spleen and para-aortic lymph nodes. Her tumor tissue was subjected to mutation analysis for epidermal growth factor receptor (EGFR) and had been shown to have a T-->G missense mutation in nucleotide 2819 of EGFR full-length cDNA (accession no. NM_005228.3), within exon 21 of the EGFR gene, resulting in amino acid substitutions from leucine to arginine at codon 858 (L858R) as expected. She received oral gefitinib 250 mg/day after mutation analysis. She had a very good tumor response with more than 90% tumor reduction shown by abdominal computed tomographic scan, normalization of previously elevated carcinoembryonic antigen level, and complete resolution of previous uptake signals in spleen and para-aortic lymph nodes shown by positron emission tomographic scan. She has been kept in clinical complete remission for 11+ months after the initiation of gefitinib treatment. Our patient supports the proposition that somatic mutation L858R in exon 21 of the EGFR gene accounts for complete responsiveness to gefitinib in a Taiwanese female patient with metastatic adenocarcinoma of lung.

18118 Background: Skin rash is the most common toxicity of EGFR-targeted therapy. Skin rash of EGFR inhibitor is associated with longer survival or tumor response. However, the clinical and genetic factors associated with this skin rash are not well understood. Methods: Fifty-two non-small-cell lung cancer patients enrolled in a prospective clinical trial of first-line gefitinib treatment were genotyped for EGFR intron 1 CA repeat polymorphism (CAn) and single nucleotide polymorphisms at promoters G-216T, C-191A, and R521K. Grade 2 to 3 skin rash within 4 weeks of treatment (early G2/3 rash) was correlated with the genotype and clinicopathological features of the patients by multivariate logistic regression. Results: Seventeen patients (32.7%) developed early G2/3 rash. In multivariate logistic regression analysis, only the CAn genotype was associated with early G2/3 rash and the effect was modified by patient age. Early G2/3 rash developed in 21% of patients with homozygous long allele (19 to 22 repeats, L) genotype (4/19), 31% of heterozygous short allele (15 to 18 repeats, S) / L genotype (8/26), and 71% of S/S genotype (5/7), respectively. The median ages of patients with early G2/3 rash and patients without early G2/3 rash were 57 years (range: 39–77) and 69 years (range: 43–86), respectively. The estimated logarithm of odds ratio (ln OR) for early G2/3 rash, as compared to S/S genotype, for S/L genotype was -0.038 multiplied by patient age (P = 0.011); and the ln OR for L/L genotype was -0.050 multiplied by patient age (P = 0.004). Fifty patients were evaluable for response. In logistic regression analysis, early G2/3 rash correlated with tumor response (P = 0.027). However, the CAn genotype was not significantly correlated with tumor response (P = 0.43). Conclusions: Homozygous short allele of EGFR CAn is more likely to develop skin rashes on gefitinib treatment. Genotyping of EGFR CAn appears to be a useful predictive marker for development of skin rashes on gefitinib use. [Table: see text]