Open Access

Improving the stability of 11C–labeled L-methionine with ascorbate

EJNMMI Radiopharmacy and Chemistry20172:13

Received: 30 July 2017

Accepted: 11 September 2017

Published: 4 October 2017



Carbon-11 labeled L-methionine (11C–MET) is a popular tracer used in the clinic for imaging brain tumors with positron emission tomography. However, the stability of 11C–MET in its final formulation is not well documented in literature. Recently, we observed fast degradation of HPLC-purified 11C–MET over time, and systematic investigation was conducted to identify the cause.


In this study, we verified the degraded product as 11C–labeled methionine sulfoxide (11C–METSO). To minimize oxidation, ascorbate (100 ppm) was added to the HPLC eluant, and the resulting HPLC-purified 11C–MET was stable in the final formulation solution without noticeable degradation for up to 1 h after the end of synthesis.


Our data suggest that to minimize degradation, ascorbate can be added to the 11C–MET formulation solution especially if it is not administered into patients soon after the end of synthesis.


C-11 L-methionine Stability Homocysteine, Methionine sulfoxide Oxidation Ascorbate Positron emission tomography


Carbon-11 labeled L-methionine (11C–MET) is a promising tracer for imaging brain tumors with positron emission tomography (PET) (Watanabe et al. 2016; Maffione et al. 2009; Glaudemans et al. 2013). Its synthesis has been previously optimized by 11C–methylation of L-homocysteine in solution or on a C-18 Sep-Pak cartridge with or without HPLC purification (Pascali et al. 1999; Tang et al. 2004; Lodia et al. 2007; Lodi et al. 2008; Boschi et al. 2009; Cheung et al. 2009; Quincoces et al. 2010; Pascali et al. 2011; Bogni et al. 2003; Nagren et al. 1998; Oh et al. 1998; Fukumura et al. 2004). While most groups reported > 97% radiochemical purity of 11C–MET at the end of synthesis (EOS) even without HPLC purification, the stability of 11C–MET in the final formulation solution has not been well documented in literature (Pascali et al. 1999; Tang et al. 2004; Lodia et al. 2007; Lodi et al. 2008; Boschi et al. 2009; Cheung et al. 2009; Quincoces et al. 2010; Pascali et al. 2011; Bogni et al. 2003; Nagren et al. 1998; Oh et al. 1998).

Bogni et al. reported the formation of a minor amount (< 6%) of C-11 L-methionine sulfoxide (11C–METSO) at 1 h after EOS in 11C–MET solution prepared either with or without HPLC purification (Fig. 1) (Bogni et al. 2003). The degradation of 11C–MET was attributed to radiolysis as the degradation rate was affected by total radioactivity and chemical composition (ethanol and L-homocysteine) of the 11C–MET solutions. Fukumura et al. observed fast degradation of HPLC-purified 11C–MET solution with radiochemical purity at 91.2 and 72.9% for samples assayed at EOS and 20 min after EOS, respectively (Fig. 1b) (Fukumura et al. 2004). The instability of 11C–MET was likely caused by radiolysis as samples with higher specific activity also showed significantly faster degradation rates. The degraded radioactive product was not identified in this report. However, it was shown that adding ethanol (EtOH, 1.5%) and Tween 80 (3%) into the final formulation solution (in saline) or using saline containing ascorbate (1000 ppm) as the eluant improved the radiochemical purity of HPLC-purified 11C–MET to 99.9%, and no significant degradation was observed over 1 h after EOS (Fukumura et al. 2004).
Fig. 1

Reported radiochemical purity at the end of synthesis and stability of 11C–MET prepared without a or with b HPLC purification

In this methodology article, we communicate our experience on the preparation of HPLC-purified 11C–MET. We report here the instability of HPLC-purified 11C–MET, our systematic investigation to find out the cause of rapid degradation, and the strategy to improve the stability of 11C–MET in the final formulation solution.

Results and discussion

To set up 11C–MET production at our institution, we used the simplest method, Sep-Pak cartridge without HPLC purification (Gomzina et al. 2011), for our initial attempt. However, we obtained much lower radiochemical purity (< 90%, data not shown) as compared to > 97% reported by others (Pascali et al. 1999; Tang et al. 2004; Lodia et al. 2007; Lodi et al. 2008; Boschi et al. 2009; Cheung et al. 2009; Quincoces et al. 2010; Pascali et al. 2011; Bogni et al. 2003). For the preparation of 11C–MET using Sep-Pak cartridge without HPLC purification, L-homocysteine and unhydrolyzed L-homocysteine thiolactone end up in the final formulation solution as well. In order to achieve higher chemical and radiochemical purities, we decided to use HPLC purification for subsequent preparations of 11C–MET.

Using phosphate-buffered saline (PBS, 3.93 mM, 3.0 mL/min) as the HPLC eluant and a C18 column (Luna C18 semi-preparative column, 5 μ, 250 × 10 mm) 11C–MET was obtained in 19 ± 4% (n = 4) decay-corrected radiochemical yield from 11CH3I with 93.5 ± 0.6% (n = 4) radiochemical purity and 21.0 ± 5.1 GBq/μmol specific activity at the end of synthesis (EOS). The analytical HPLC column was a Luna C18 column (5 μ, 250 × 4.6 mm), the eluant was phosphate buffer (1 mM, pH 3), and the flow rate was 1.0 mL/min. A minor radioactive by-product was observed at t R = ~3.3 min (Fig. 2a). No residual L-homocysteine or L-homocysteine thiolactone was detected in the final 11C–MET solution as monitored by UV detector (set at 220 nm). The radiochemical purity of 11C–MET quickly dropped to 75 ± 6% (n = 4) at 1 h after EOS with concurrent increase of the radioactive by-product (Fig. 2b). Our radiochemical purity data (93.5% at EOS and 75% at 1 h after EOS) were lower than the data (> 99% at EOS and > 96% at 1 h after EOS) reported by Bogni et al. (Bogni et al. 2003), but were comparable to the data (91.2% at EOS and 72.9% at 20 min after EOS) reported by Fukumura et al. (Fukumura et al. 2004).
Fig. 2

Representative radio-HPLC chromatograms of 11C–MET (t R = ~5.6 min) purified by HPLC using PBS as the eluant: a assayed at EOS and b assayed at 1 h after EOS

To our knowledge, there was only one report on degradation of 11C–MET prepared using Sep-Pak cartridge without HPLC purification (Bogni et al. 2003). A minor amount (< 4%) of 11C–METSO was formed over a 1-h period after EOS (Bogni et al. 2003). Presumably, the remaining L-homocysteine ending up in the final formulation solution could serve as a free radical scavenger, and prevent degradation of 11C–MET. To verify this hypothesis, we added 0.5 mg of L-homocysteine in the product collection vial of the synthesis module. After mixing with the HPLC-purified 11C–MET eluate fraction, the mixed solution was passed through a sterile filter and checked by HPLC. As shown in Fig. 3a, the radiochemical purity of 11C–MET increased to 97.5 ± 1.5% (n = 2) at EOS with a minor radioactive by-product eluted at the same retention (t R = ~3.3 min) of the degraded product as shown in Fig. 2b. However, the content of this radioactive by-product still slowly increased over time even in the presence of L-homocysteine (radiochemical purity of 11C–MET: 91.1 ± 8.3% at 1 h after EOS, Fig. 3b).
Fig. 3

Representative Radio-HPLC chromatograms of 11C–MET purified by HPLC using PBS as the eluant and spiked with homocysteine: a assayed at EOS and b assayed at 1 h after EOS

Next, we tried to identify the radioactive degradation product in our HPLC-purified 11C–MET solution. Previously, 11C–METSO was reported to be the degradation product of 11C–MET prepared with or without HPLC purification (Bogni et al. 2003). To verify this, we co-injected the degraded 11C–MET solution (1 h after EOS) with MET and METSO into HPLC. As shown in Fig. 4, the UV peak of METSO co-eluted with the radio peak of the radioactive by-product (t R = ~3.3 min), confirming 11C–METSO was indeed the degradation by-product of our HPLC-purified 11C–MET. In addition, we also conducted mass analysis for the decayed product solution, and confirmed the presence of METSO (see Additional file 1: Figure S1).
Fig. 4

HPLC chromatograms of co-injecting MET and METSO with 11C–MET solution purified by HPLC using PBS as the eluant: a UV chromatogram b Radio chromatogram

Previously, Bogni et al. suggested that 11C–MET prepared without HPLC purification could be stabilized by the remaining EtOH and L-homocysteine in the final product solution (Bogni et al. 2003). Fukumura et al. also demonstrated that HPLC-purified 11C–MET could be stabilized by the addition of EtOH (1.5%) and Tween 80 (3%) to the final product solution (Fukumura et al. 2004). Since EtOH is relatively nontoxic and readily available, we tested if EtOH alone could be used to stabilize the HPLC-purified 11C–MET solution. We added EtOH (4%) immediately after the 11C–MET-containing HPLC eluate fraction was collected in the final product vial, and checked radiochemical purity of 11C–MET over time. The radiochemical purities of EtOH-containing 11C–MET solution were 94.7 ± 3.3% and 77.7 ± 16.1% (n = 2) at EOS and 1 h after EOS, respectively. Representative HPLC chromatograms are shown in Fig. 5. These data suggest that EtOH alone is not very effective to prevent degradation of HPLC-purified 11C–MET solution. The previous findings by others might be due to the combination of EtOH with L-homocysteine or Tween 80 (Bogni et al. 2003; Fukumura et al. 2004).
Fig. 5

Radio-HPLC chromatograms of 11C–MET purified by HPLC using PBS as the eluant and spiked with EtOH (4%): a assayed at EOS and b assayed at 1 h after EOS

Interestingly, 11C–METSO was observed in Figs. 2, 3 and 5 even at EOS, suggesting 11C–METSO was formed quickly after 11C–MET was separated from L-homocysteine during HPLC purification. To stabilize 11C–MET final formulation solution and minimize the formation of 11C–METSO even during the HPLC purification process, we added ascorbate directly into HPLC solvent. Instead of 1000 ppm tested by Fukumura et al. (Fukumura et al. 2004), we used only 100 ppm of ascorbate. The eluate fraction containing 11C–MET was collected and passed through a sterile filter. 11C–MET was obtained in 22 ± 3% (n = 8) decay-corrected radiochemical yield from 11CH3I with a 99.2 ± 0.9% (n = 8) radiochemical purity at EOS. No significant degradation of 11C–MET solution was observed as the radiochemical purity was 98.2 ± 1.7% (n = 8) at 1 h after EOS (Fig. 6). These data are consistent with the observation of Fukumura et al. (Fukumura et al. 2004), and suggest that 100 ppm of ascorbate is sufficient to minimize the formation of 11C–METSO.
Fig. 6

Representative radio-HPLC chromatograms of 11C–MET purified by HPLC using PBS containing 100 ppm of ascorbate as the eluant: a assayed at EOS and b assayed at 1 h after EOS


We successfully verified the degradation of HPLC-purified 11C–MET was due to the formation of 11C–METSO. Presence of L-homocysteine or EtOH in the final 11C–MET formulation solution could slow down the degradation of 11C–MET. Adding ascorbate to the HPLC solvent greatly improved the radiochemical purity and stability of HPLC-purified 11C–MET solution. This could be very useful especially if 11C–MET is not used immediately after EOS. The tested concentration (100 ppm) contains only ~1.4 mg of ascorbate in the entire dose (~13.5 mL). It is safe for administration as this mass of ascorbate is much lower than the usual therapeutic parenteral dose (100–250 mg).


General methods

METSO and sodium phosphate monobasic were purchased from Sigma-Aldrich (Oakville, Canada). For preparation of the QC HPLC solvent and the phosphate buffer used to elute the reaction mixture off the cartridge, sodium phosphate monobasic was diluted with water to the specified concentrations without adjusting pH. Vitamin C injection (ascorbic acid, 250 mg/mL) and sodium phosphate injection (3 mmol/mL) were purchased from Sandoz (Boucherville, Canada). Saline injection (0.9% NaCl) was purchased from Baxter (Mississauga, Canada). The semi-preparative HPLC solvents were prepared by mixing sodium phosphate injection and saline injection (with or without vitamin C injection) to the specified concentrations. All other chemicals and solvents were obtained from commercial sources, and used without further purification. Sep-Pak tC18 Plus Short cartridges (400 mg) were obtained from Waters (Milford, MA). C-11 methane was produced by 18-MeV proton bombardment of an N2/H2 (10% H2 in N2) target using an Advanced Cyclotron Systems Inc. (Richmond, Canada) TR19 cyclotron. C-11 methane was converted to C-11 methyl iodide (11CH3I) in gas phase using a GE (Chicago, IL) TRACER FX C Pro module. Purification of 11C–MET was conducted using the HPLC component of the synthesis module on a Phenomenex (Torrance, CA) Aqua C18 semi-preparative column (5 μ, 250 × 10 mm). The HPLC solvent was phosphate-buffered saline (PBS, 3.93 mM) or PBS containing 100 ppm of ascorbate, and the flow rate was 3.0 mL/min. Millex-GS 0.22 μm sterile filter was purchased from EMD Millipore (Billerica, MA). Radioactivity was measured using a Capintec (Ramsey, NJ) CRC®-Ultra R dose calibrator. Mass analysis was performed using an AB SCIEX (Framingham, MA, USA) 4000 QTRAP mass spectrometer system with an ESI ion source.

Synthesis and purification of 11C–MET

The tC18 Sep-Pak cartridge was preconditioned with EtOH (5 mL) and sterile water (10 mL). The remaining water in the cartridge was pushed out with air (10 mL). Five minutes before EOS, 85 μL of L-homocysteine thiolactone hydrochloride aqueous solution (25 mg in 600 μL water) was mixed with 200 μL of NaOH solution (0.7 mL of 10 N NaOH aqueous solution diluted with 4.3 mL water and 5.0 mL EtOH). From this, 200 μL of the mixed solution was loaded to the tC18 Sep-Pak cartridge. After passing 11CH3I by helium (15 mL/min) through the tC18 Sep-Pak cartridge, the reaction mixture was eluted off the cartridge with phosphate buffer (50 mM, 2 mL) and purified by HPLC. The eluate fraction (~ 1.5 mL) containing 11C–MET was collected, diluted with HPLC eluant (12 mL), and passed through a Millex-GS sterile filter into a final product vial.

Quality control of 11C–MET

Chemical purity, radiochemical purity and radiochemical identity of 11C–MET and by-products were determined using an Agilent (Santa Clara, CA) HPLC system equipped with a model 1200 quaternary pump, a model 1200 UV absorbance detector (set at 220 nm), and a Bioscan (Washington, DC) NaI scintillation detector. The operation of the Agilent HPLC system was controlled using the Agilent ChemStation software. The HPLC column used was a Phenomenex Luna C18 analytical column (5 μ, 250 × 4.6 mm). The HPLC solvent was phosphate buffer (1 mM, pH 3), and the flow rate was 1.0 mL/min.



Carbon-11 labeled methyl iodide


Carbon-11 labeled L-methionine


Carbon-11 labeled methionine sulfoxide


End of synthesis




High performance liquid chromatography


Sodium hydroxide


Phosphate-buffered saline


Positron emission tomography


Part per million

t R

Retention time


Ultraviolet light



This work was supported by the Canadian Institutes of Health Research (FDN-148465) and the Leading Edge Endowment Fund.

Authors’ contributions

MW, DW, FB and KSL conceived and designed the complete study. MW, LL, KF, JGG, ZZ, CZ and WE conducted the experiments. MW, LL, ZZ, CZ and KSL summarized and interpreted the data. The manuscript was drafted by KSL with critical revisions from MW, LL, KF, JGG, ZZ, CZ, and WE. All authors read and approved the final manuscript.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors’ Affiliations

Department of Functional Imaging, BC Cancer Agency
Department of Molecular Oncology, BC Cancer Agency
Department of Radiology, University of British Columbia


  1. Bogni A, Bombardieri E, Iwata R, Cadini L, Pascali C. Stability of L-[S-methyl-11C]methionine solutions. J Radioanal Nucl Chem. 2003;256:199–203.View ArticleGoogle Scholar
  2. Boschi S, Lodi F, Cicoria G, Ledesma JR, Knopp R, Rizzello A, Di Pierro D, Trespidi S, Marengo M. Development of a modular system for the synthesis of PET [11C]labelled radiopharmaceuticals. Appl Radiat Isot. 2009;67:1869–73.View ArticlePubMedGoogle Scholar
  3. Cheung M-K, Ho C-L. A simple, versatile, low-cost and remotely operated apparatus for [11C]acetate, [11C]choline, [11C]methionine and [11C]PIB synthesis. Appl Radiat Isot. 2009;67:581–9.View ArticlePubMedGoogle Scholar
  4. Fukumura T, Nakao R, Yamaguchi M, Suzuki K. Stability of 11C-labeled PET radiopharmaceuticals. Appl Radiat Isot. 2004;61:1279–87.View ArticlePubMedGoogle Scholar
  5. Glaudemans AWJM, Enting RH, Heesters MAAM, Dierckx RAJO, van Rheenen RWJ, Walenkamp AME, Slart RHJA. Value of 11C-methionine PET in imaging brain tumours and metastases. Eur J Nucl Med Mol Imaging. 2013;40:615–35.View ArticlePubMedGoogle Scholar
  6. Gomzina NA, Kuznetsova OF. L-[Methyl-(11C)]-methionine of high enantiomeric purity production via online-11C-methylation of L-homocysteine thiolactone hydrochloride. Russ J Bioorg Chem. 2011;37:191–7.View ArticleGoogle Scholar
  7. Lodi F, Rizzello A, Trespidi S, Di Pierro D, Marengo M, Farsad M, Fanti S, Al-Nahhas A, Rubello D, Boschi S. Reliability and reproducibility of N-[11C]methyl-choline and L-(S-methyl-[11C])methionine solid-phase synthesis: a useful and suitable method in clinical practice. Nucl Med Commun. 2008;29:736–40.View ArticlePubMedGoogle Scholar
  8. Lodi F, Trespidi S, Di Pierro D, Marengo M, Farsad M, Fanti S, Franchi R, Boschi S. A simple Tracerlab module modification for automated on-column [11C]methylation and [11C]carboxylation. Appl Radiat Isot. 2007;65:691–5.View ArticlePubMedGoogle Scholar
  9. Maffione AM, Nanni C, Ambrosini V, Trespidi S, Lopci E, Allegri V, Castellucci P, Montini G, Boschi S, Fanti S. 11C-Methionine PET/CT in central nervous system tumours: a review. Curr Radiopharma. 2009;2:160–4.View ArticleGoogle Scholar
  10. Nagren K, Halldin C. Methylation of amide and thiol functions with [11C]methyl triflate, as exemplified by [11C]NMSP, [11C]flumazenil and [11C]methionine. J Label Compd Radiopharm. 1998;41:831–41.View ArticleGoogle Scholar
  11. Oh S-J, Choe YS, Kim YS, Choi Y, Kim SE, Lee KH, Ha H-J, Kim B-T. Development of an automated system for the routine preparation of carbon-11 labeled radiopharmaceuticals. Bull Kor Chem Soc. 1998;19:952–6.Google Scholar
  12. Pascali C, Bogni A, Cucchi C, Laera L, Crispu O, Maiocchi G, Crippa F, Bombardieri E. Detection of additional impurities in the UV-chromatogram of L-[S-methyl-11C]methionine. J Radioanal Nucl Chem. 2011;288:405–9.View ArticleGoogle Scholar
  13. Pascali C, Bogni A, Iwata R, Decise D, Crippa F, Bombardieri E. High efficiency preparation of L-[S-methyl-11C]methionine by on-column [11C]methylation on C18 sep-Pak. J Label Compd Radiopharm. 1999;42:715–24.View ArticleGoogle Scholar
  14. Quincoces G, Lopez-Sanchez L, Sanchez-Martınez M, Rodrıguez-Fraile M, Penuelas I. Design and performance evaluation of single-use whole-sterile “plug & play” kits for routine automated production of [11C]choline and [11C]methionine with radiopharmaceutical quality. Appl Radiat Isot. 2010;68:2298–301.View ArticlePubMedGoogle Scholar
  15. Tang G-H, Wang M-F, Tang X-L, Luo L, Gan M-Q. Automated synthesis of (S-[11C]-methyl)-L-methionine and (S-[11C]-methyl)-L-cycteine by on-column [11C]methylation. J Nucl Radiochem. 2004;26:77–83.Google Scholar
  16. Watanabe A, Muragaki Y, Maruyama T, Shinoda J, Okada Y. Usefulness of 11C-methionine positron emission tomography for treatment-decision making in cases of non-enhancing glioma-like brain lesions. J Neuro-Oncol. 2016;126:577–83.View ArticleGoogle Scholar


© The Author(s) 2017