Metabolite Biomarkers of Prolonged and Intensified Pain and Distress in Head and Neck Cancer Patients Undergoing Radio- or Chemoradiotherapy by Means of NMR-Based Metabolomics-A Preliminary Study

Abstract

Treatment of head and neck squamous cell carcinoma (HNSCC) has a detrimental impact on patient quality of life. The rate of recognized distress/depression among HNSCC patients ranges from 9.8% to 83.8%, and the estimated prevalence of depression among patients receiving radiotherapy is 63%. Shorter overall survival also occurs in preexisting depression or depressive conditions. The present study analyzes the nuclear magnetic resonance (NMR) blood serum metabolic profiles during radio-/chemoradiotherapy and correlates the detected alterations with pain and/or distress accumulated with the disease and its treatment. NMR spectra were acquired on a Bruker 400 MHz spectrometer and analyzed using multivariate methods. The results indicate that distress and/or pain primarily affect the serum lipids and metabolites of energy (glutamine, glucose, lactate, acetate) and one-carbon (glycine, choline, betaine, methanol, threonine, serine, histidine, formate) metabolism. Sparse disturbances in the branched-chain amino acids (BCAA) and in the metabolites involved in protein metabolism (lysine, tyrosine, phenylalanine) are also observed. Depending on the treatment modality-radiotherapy or concurrent chemoradiotherapy-there are some differences in the altered metabolites.

Keywords: HNSCC; chemotherapy; distress; metabolomics; pain; radiotherapy.

Figure 1

The OPLS-DA s-line plots with the indicated metabolites responsible for discrimination between the L (Low) and H (High) WOD classes for CHRT+RT (a), RT (b) and CHRT (c). The numbering of the metabolites corresponds to that used in Table 4. 1—lipids, 2—glutamine, 4—lactate, 6—glycine, 9—methanol, 11—serine, 14—lysine, 15—tyrosine, 16—phenylalanine, 18—valine. The corresponding cross-validated score plots are presented in Figure S2.

Figure 2

The OPLS-DA s-line plots with the indicated metabolites responsible for discrimination between the L (Low) and H (High) MD classes for CHRT+RT (a), RT (b) and CHRT (c). The numbering of the metabolites corresponds to that used in Table 4. 1—lipids, 2—glutamine, 3—glucose, 10—threonine, 11—serine. The corresponding cross-validated score plots are presented in Figure S3.

Figure 3

The OPLS-DA s-line plots with the indicated metabolites responsible for discrimination between the L (Low) and H (High) WOP classes for CHRT+RT (a), RT (b) and CHRT (c). The numbering of the metabolites corresponds to that used in Table 4. 1—lipids, 2—glutamine, 3—glucose, 4—lactate, 5—acetate, 6—glycine, 7—choline, 8—betaine, 9—methanol, 10—threonine, 11—serine, 12—histidine, 16—phenylalanine, 17—leucine, 18—valine, 19—isoleucine. The corresponding cross-validated score plots are presented in Figure S4.

Figure 4

The OPLS-DA s-line plots with the indicated metabolites responsible for discrimination between the L (Low) and H (High) MP classes for CHRT+RT (a), RT (b) and CHRT (c). The numbering of the metabolites corresponds to that used in Table 4. 1—lipids, 2—glutamine, 3—glucose, 8—betaine, 15—tyrosine, 16—phenylalanine. In contrary to the previous analyses, a higher MP is characterized by the decreased lipid signals. The corresponding cross-validated score plots are presented in Figure S5.

Figure 5

Comparison of metabolic alterations in the analyzed subgroups (WOD, MD, WOP and MD) for RT and CHRT. The length of bar corresponds to the value of the p(corr) parameter from OPLS-DA.