Increasing Dosage of Leucovorin Results in Pharmacokinetic and Gene Expression Differences When Administered as Two-Hour Infusion or Bolus Injection to Patients with Colon Cancer

Abstract

The combination of 5-fluorouracil (5-FU) and leucovorin (LV) forms the chemotherapy backbone for patients with colorectal cancer. However, the LV administration is often standardized and not based on robust scientific data. To address these issues, a randomized pharmacokinetics study was performed in patients with colon cancer. Thirty patients were enrolled, receiving 60, 200 or 500 mg/m² LV as a single two-hour infusion. Blood, tumor, mucosa, and resection margin biopsies were collected. Folate concentrations were analyzed with LC-MS/MS and gene expression with qPCR. Data from a previous study where patients received LV as bolus injections were used as comparison. Saturation of methylenetetrahydrofolate (MeTHF) and tetrahydrofolate (THF) levels was seen after two-hour infusion and polyglutamated MeTHF + THF levels in tumors decreased with increasing LV dosage. The decrease was associated with decreased FPGS and increased GGH expression, which was not observed after LV bolus injection. In the bolus group, results indicate activation of a metabolic switch possibly promoting TYMS inhibition in response to 5-FU. Different metabolic mechanisms appear to be induced when LV is administered as infusion and bolus injection. Since maximal inhibition of TYMS by the 5-FU metabolite 5-fluoro-2'-deoxyuridine 5'-monophosphate (FdUMP) requires excess polyglutamated MeTHF, the results point in favor of the bolus regimen.

Keywords: LC-MS/MS; bolus injection; colon cancer; gene expression; leucovorin; methylenetetrahydrofolate; pharmacokinetics; two-hour infusion.

Figure 1

Simplified overview showing compartmentalization of folate metabolism in eukaryotic cells. Monoglutamated folates (Fol⁻), including leucovorin (LV), enter cells through RFC-1, PCFT, or FRs. Inside cells, polyglutamation of folates occur through the action of FPGS, whereas GGH yields monoglutamated folates. These may be exported out of cells by ABCC3. In the cytosol, LV may be enzymatically converted to MeTHF by two different routes: either by MTHFS and MTHFD1, or by MTHFS, MTHFD1, ALDH1L1, and SHMT1. Similar reactions occur in the mitochondria, where LV can be converted to MeTHF by MTHFS and MTHFD2, or by MTHFS, MTHFD2, ALDH1L2, and SHMT2. De novo dTMP synthesis in the cytosol occurs through the activities of the enzymes SHMT1, TYMS, and DHFR, and in mitochondria through SHMT2, TYMS, and DHFRL1 localized to the mitochondrial matrix and inner membrane. In the nucleus, sumoylated forms of SHMT1, TYMS, and DHFR forms a complex needed for nuclear folate metabolism. Abbreviations: ABCC3: ATP-binding cassette, subfamily C (CFTR/MRP), member 3; ALDH1L1/2: aldehyde dehydrogenase 1 family member L1/2; DHF: dihydrofolate; DHFR: dihydrofolate reductase; dUMP: deoxyuridine monophosphate; dTMP: deoxythymidine monophosphate; FPGS: folylpolyglutamate synthase; Fol⁻: monoglutamated folate; FR: folate receptor; GGH: gamma-glutamyl hydrolase; MeTHF: methylenetetrahydrofolate; MTHFD1L: methylenetetrahydrofolate dehydrogenase (NADP+ dependent) 1-like; MTHFD2:methylenetetrahydrofolate dehydrogenase (NAPD+ dependent) 2; MTHFS: 5,10-methenyltetrahydrofolate synthetase; OP⁻: organic phosphate; PCFT: solute carrier family 46 (folate transporter), member 1, proton coupled folate transporter; RFC-1: solute carrier family 19 (folate transporter) member 1; SHMT1/2/2α: serine hydroxymethyltransferase 1/2/2α; TYMS: thymidylate synthase; THF: tetrahydrofolate.

Figure 2

Dose-AUC relationship among patients receiving a two-hour intravenous infusion of 60, 200, or 500 mg/m² LV. The figure (a) shows the AUC (pmol/mL × 24 h) of LV and the figure (b) shows the AUC of MTHF in plasma of the 30 included patients. Due to the large difference in LV and MTHF levels, the scale is based on the natural logarithm.

Figure 3

Relationship between (a) tissue LV level and plasma LV AUC, (b) tissue (MeTHF + THF) level and plasma LV AUC, and (c) tissue (MeTHF + THF) and LV levels, at different two-hour infusion dosages of LV. * p ≤ 0.05; ** p ≤ 0.01, *** p ≤ 0.001.

Figure 4

Box plots showing median gene expression levels in tumors (t) and mucosa (m) of colon cancer patients at baseline (0, n = 10) and after administration of 60 mg/m² (Inf n = 10, Bol n = 10), 200 mg/m² (Inf n = 10, Bol n = 8), or 500 mg/m² (Inf n = 10, Bol n = 10) LV administered as two-hour infusion (Inf, shaded boxes) or bolus (Bol) injections. Significance asterisks above boxes show differences in tumors or mucosa related to LV dosage whereas significance asterisks below boxes show differences in tumors or mucosa relating to LV administration (two-hour infusion or bolus injection) at a specific LV dosage. * p ≤ 0.05; ** p ≤ 0.01, *** p ≤ 0.001.

Figure 5

Heat maps showing the correlation between gene expression and folate levels in (a) tumors (t) and (b) mucosa (m) at baseline. The strongest positive correlation (1.0) is shown in red whereas the strongest negative correlation (−1.0) is shown in purple. LV in tissue was not analyzed at baseline.

Figure 6

Heat maps showing the correlation between gene expression and folate levels in tumor and mucosa tissues after administration of different dosages of LV given either as two-hour infusion (left panels) or bolus injection (right panels). The strongest positive correlation (1.0) is shown in red whereas the strongest negative correlation (−1.0) is shown in purple. LV in tissue was not analyzed in the bolus group.

Figure 7

Ratio of polyglutamated/monoglutamated folates, FPGS gene expression, and GGH gene expression in tumors (a,c,e) and mucosa (b,d,f) of patients receiving no LV, as well as different doses of LV given as two-hour infusion (blue boxes) or bolus injection (red boxes). * p ≤ 0.05; ** p ≤ 0.01, *** p ≤ 0.001.