Summary of Study ST002082
This data is available at the NIH Common Fund's National Metabolomics Data Repository (NMDR) website, the Metabolomics Workbench, https://www.metabolomicsworkbench.org, where it has been assigned Project ID PR001322. The data can be accessed directly via it's Project DOI: 10.21228/M8TT44 This work is supported by NIH grant, U2C- DK119886.
See: https://www.metabolomicsworkbench.org/about/howtocite.php
This study contains a large results data set and is not available in the mwTab file. It is only available for download via FTP as data file(s) here.
| Study ID | ST002082 |
| Study Title | Predicting dying: a study of the metabolic changes and the dying process in patients with lung cancer |
| Study Type | Observational study |
| Study Summary | Background: Accurately recognising that a person may be dying is central for improving their experience of care. Yet recognising dying is difficult and predicting dying frequently inaccurate. Methods: Serial urine samples from patients (n=112) with lung cancer were analysed using high resolution untargeted mass spectrometry. ANOVA and volcano plot analysis demonstrated metabolites that changed in the last weeks of life. Further analysis identified potential biological pathways affected. Cox lasso logistic regression was engaged to develop a multivariable model predicting the probability of survival within the last 30 days of life. Results: In total 124 metabolites changed. ANOVA analysis identified 93 metabolites and volcano plot analysis 85 metabolites. 53 metabolites changed using both approaches. Pathways altered in the last weeks included those associated with decreased oral intake, muscle loss, decreased RNA and protein synthesis, mitochondrial dysfunction, disrupted β-oxidation and one carbon metabolism. Epinephrine and cortisol increased in the last 2 weeks and week respectively. A model predicting time to death using 7 metabolites had excellent accuracy (AUC= 0.86 at day 30, 0.88 at day 20 and 0.85 at day 10) and enabled classification of patients at low, medium and high risk of dying on a Kaplan-Meier survival curve. Conclusions: Metabolomic analysis identified metabolites and their associated pathways that change in the last weeks and days of life in patients with lung cancer. Prognostic tests based on the metabolites identified have the potential to change clinical practice and improve the care of dying patients. |
| Institute | University of Liverpool Institute of Life Course & Medical Sciences |
| Last Name | Norman |
| First Name | Brendan |
| Address | William Henry Duncan Building, 6 West Derby Street, Liverpool, UK. L7 8TX |
| bnorman@liverpool.ac.uk | |
| Phone | (+44)151 794 9064 |
| Submit Date | 2022-01-24 |
| Num Groups | 6 |
| Total Subjects | 112 |
| Num Males | 67 |
| Num Females | 45 |
| Raw Data Available | Yes |
| Raw Data File Type(s) | d |
| Analysis Type Detail | LC-MS |
| Release Date | 2022-02-24 |
| Release Version | 1 |
Select appropriate tab below to view additional metadata details:
Combined analysis:
| Analysis ID | AN003396 | AN003397 | AN003398 | AN003399 |
|---|---|---|---|---|
| Analysis type | MS | MS | MS | MS |
| Chromatography type | Reversed phase | Reversed phase | HILIC | HILIC |
| Chromatography system | Agilent 6550 | Agilent 6550 | Agilent 6550 | Agilent 6550 |
| Column | Waters Atlantis dC18 (100 x 3mm,3um) | Waters Atlantis dC18 (100 x 3mm,3um) | Waters BEH Amide (150 x 3.0mm,1.7um) | Waters BEH Amide (150 x 3.0mm,1.7um) |
| MS Type | ESI | ESI | ESI | ESI |
| MS instrument type | QTOF | QTOF | QTOF | QTOF |
| MS instrument name | Agilent 6550 QTOF | Agilent 6550 QTOF | Agilent 6550 QTOF | Agilent 6550 QTOF |
| Ion Mode | NEGATIVE | POSITIVE | NEGATIVE | POSITIVE |
| Units | Values are raw peak area | raw peak area | raw peak area | Values are raw peak area |
MS:
| MS ID: | MS003163 |
| Analysis ID: | AN003396 |
| Instrument Name: | Agilent 6550 QTOF |
| Instrument Type: | QTOF |
| MS Type: | ESI |
| MS Comments: | Mass spectrometry conditions: The mass spectrometer was tuned and calibrated according to protocols recommended by the manufacturer. Acquisition was performed in 2 GHz mode and mass range 50-1700. The capillary voltage was 4000 V and fragmentor voltage 380 V. The desolvation gas temperature was 200 °C with flow rate at 15 L/min. The sheath gas temperature was 300 °C with flow rate at 12 L/min. The nebulizer pressure was 40 psig and nozzle voltage 1000 V (± for positive and negative ionisation modes). The acquisition rate was 3 spectra/second. The reference mass solution was continually infused at a flow rate of 0.5 mL/min by a separate isocratic pump for constant mass correction. Repeat injections of each pooled sample were interspersed throughout the analytical sequence, as per the quality control procedure described by Norman et al. (Clin Chem. 2019;65(4):530-539. doi: 10.1373/clinchem.2018.295345). The individual sample analysis order was randomised computationally. Data Pre-processing and Quality Control: All data were acquired using the MassHunter suite (Agilent build 6.0) with quality checks being performed by Qualitative Analysis (build 07.00). Mass accuracy was checked using extracted ion chromatograms of reference masses: the resulting accuracy was ±5 ppm during the run. Additionally, chromatographic reproducibility was checked by overlaying binary pump pressure curves across each analytical sequence. Data was filtered based upon the pooled QC samples, with compounds being retained if observed in 100 % of replicate injections for at least 1 pool and with peak area coefficient of variation (CV) <25 % across all replicate injections for each pool. A comprehensive semi-targeted approach was employed to assign the identity of urinary metabolites using an in-house compound library that included a broad range of metabolites involved in intermediary metabolism. Targeted feature extraction was performed on each dataset based on matching of metabolite chemical features against an accurate mass and retention time (AMRT) database previously generated from analysis of the IROA Technology MS metabolite library of tandards by each LC method described above, combined with the same QTOF analytical parameters used in this study (databases publicly available: https://doi.org/10.6084/m9.figshare.c.4378235.v2). In addition to accurate mass and retention time, MS/MS (data-dependent, employed for 'hit' metabolites) was also used in the confirmation of metabolite identity (i.e. level 1 identification) as per Sumner et al. Feature extraction was performed in MassHunter Profinder (build 10.0); accurate mass window of 10 ppm and retention time window of 0.3 min against the respective AMRT database. Data were exported in the format of a csv file for statistical analysis. Unidentified compounds are 'known unknowns' of interest from previous experiments; detection by matched AMRT and named in the format 'RT_neutral-mass'. 'BDRM'; unidentified bone-derived metabolite. |
| Ion Mode: | NEGATIVE |
| MS ID: | MS003164 |
| Analysis ID: | AN003397 |
| Instrument Name: | Agilent 6550 QTOF |
| Instrument Type: | QTOF |
| MS Type: | ESI |
| MS Comments: | Mass spectrometry conditions: The mass spectrometer was tuned and calibrated according to protocols recommended by the manufacturer. Acquisition was performed in 2 GHz mode and mass range 50-1700. The capillary voltage was 4000 V and fragmentor voltage 380 V. The desolvation gas temperature was 200 °C with flow rate at 15 L/min. The sheath gas temperature was 300 °C with flow rate at 12 L/min. The nebulizer pressure was 40 psig and nozzle voltage 1000 V (± for positive and negative ionisation modes). The acquisition rate was 3 spectra/second. The reference mass solution was continually infused at a flow rate of 0.5 mL/min by a separate isocratic pump for constant mass correction. Repeat injections of each pooled sample were interspersed throughout the analytical sequence, as per the quality control procedure described by Norman et al. (Clin Chem. 2019;65(4):530-539. doi: 10.1373/clinchem.2018.295345). The individual sample analysis order was randomised computationally. Data Pre-processing and Quality Control: All data were acquired using the MassHunter suite (Agilent build 6.0) with quality checks being performed by Qualitative Analysis (build 07.00). Mass accuracy was checked using extracted ion chromatograms of reference masses: the resulting accuracy was ±5 ppm during the run. Additionally, chromatographic reproducibility was checked by overlaying binary pump pressure curves across each analytical sequence. Data was filtered based upon the pooled QC samples, with compounds being retained if observed in 100 % of replicate injections for at least 1 pool and with peak area coefficient of variation (CV) <25 % across all replicate injections for each pool. A comprehensive semi-targeted approach was employed to assign the identity of urinary metabolites using an in-house compound library that included a broad range of metabolites involved in intermediary metabolism. Targeted feature extraction was performed on each dataset based on matching of metabolite chemical features against an accurate mass and retention time (AMRT) database previously generated from analysis of the IROA Technology MS metabolite library of tandards by each LC method described above, combined with the same QTOF analytical parameters used in this study (databases publicly available: https://doi.org/10.6084/m9.figshare.c.4378235.v2). In addition to accurate mass and retention time, MS/MS (data-dependent, employed for 'hit' metabolites) was also used in the confirmation of metabolite identity (i.e. level 1 identification) as per Sumner et al. Feature extraction was performed in MassHunter Profinder (build 10.0); accurate mass window of 10 ppm and retention time window of 0.3 min against the respective AMRT database. Data were exported in the format of a csv file for statistical analysis. Unidentified compounds are 'known unknowns' of interest from previous experiments; detection by matched AMRT and named in the format 'RT_neutral-mass'. 'BDRM'; unidentified bone-derived metabolite. |
| Ion Mode: | POSITIVE |
| MS ID: | MS003165 |
| Analysis ID: | AN003398 |
| Instrument Name: | Agilent 6550 QTOF |
| Instrument Type: | QTOF |
| MS Type: | ESI |
| MS Comments: | Mass spectrometry conditions: The mass spectrometer was tuned and calibrated according to protocols recommended by the manufacturer. Acquisition was performed in 2 GHz mode and mass range 50-1700. The capillary voltage was 4000 V and fragmentor voltage 380 V. The desolvation gas temperature was 200 °C with flow rate at 15 L/min. The sheath gas temperature was 300 °C with flow rate at 12 L/min. The nebulizer pressure was 40 psig and nozzle voltage 1000 V (± for positive and negative ionisation modes). The acquisition rate was 3 spectra/second. The reference mass solution was continually infused at a flow rate of 0.5 mL/min by a separate isocratic pump for constant mass correction. Repeat injections of each pooled sample were interspersed throughout the analytical sequence, as per the quality control procedure described by Norman et al. (Clin Chem. 2019;65(4):530-539. doi: 10.1373/clinchem.2018.295345). The individual sample analysis order was randomised computationally. Data Pre-processing and Quality Control: All data were acquired using the MassHunter suite (Agilent build 6.0) with quality checks being performed by Qualitative Analysis (build 07.00). Mass accuracy was checked using extracted ion chromatograms of reference masses: the resulting accuracy was ±5 ppm during the run. Additionally, chromatographic reproducibility was checked by overlaying binary pump pressure curves across each analytical sequence. Data was filtered based upon the pooled QC samples, with compounds being retained if observed in 100 % of replicate injections for at least 1 pool and with peak area coefficient of variation (CV) <25 % across all replicate injections for each pool. A comprehensive semi-targeted approach was employed to assign the identity of urinary metabolites using an in-house compound library that included a broad range of metabolites involved in intermediary metabolism. Targeted feature extraction was performed on each dataset based on matching of metabolite chemical features against an accurate mass and retention time (AMRT) database previously generated from analysis of the IROA Technology MS metabolite library of tandards by each LC method described above, combined with the same QTOF analytical parameters used in this study (databases publicly available: https://doi.org/10.6084/m9.figshare.c.4378235.v2). In addition to accurate mass and retention time, MS/MS (data-dependent, employed for 'hit' metabolites) was also used in the confirmation of metabolite identity (i.e. level 1 identification) as per Sumner et al. Feature extraction was performed in MassHunter Profinder (build 10.0); accurate mass window of 10 ppm and retention time window of 0.3 min against the respective AMRT database. Data were exported in the format of a csv file for statistical analysis. |
| Ion Mode: | NEGATIVE |
| MS ID: | MS003166 |
| Analysis ID: | AN003399 |
| Instrument Name: | Agilent 6550 QTOF |
| Instrument Type: | QTOF |
| MS Type: | ESI |
| MS Comments: | Mass spectrometry conditions: The mass spectrometer was tuned and calibrated according to protocols recommended by the manufacturer. Acquisition was performed in 2 GHz mode and mass range 50-1700. The capillary voltage was 4000 V and fragmentor voltage 380 V. The desolvation gas temperature was 200 °C with flow rate at 15 L/min. The sheath gas temperature was 300 °C with flow rate at 12 L/min. The nebulizer pressure was 40 psig and nozzle voltage 1000 V (± for positive and negative ionisation modes). The acquisition rate was 3 spectra/second. The reference mass solution was continually infused at a flow rate of 0.5 mL/min by a separate isocratic pump for constant mass correction. Repeat injections of each pooled sample were interspersed throughout the analytical sequence, as per the quality control procedure described by Norman et al. (Clin Chem. 2019;65(4):530-539. doi: 10.1373/clinchem.2018.295345). The individual sample analysis order was randomised computationally. Data Pre-processing and Quality Control: All data were acquired using the MassHunter suite (Agilent build 6.0) with quality checks being performed by Qualitative Analysis (build 07.00). Mass accuracy was checked using extracted ion chromatograms of reference masses: the resulting accuracy was ±5 ppm during the run. Additionally, chromatographic reproducibility was checked by overlaying binary pump pressure curves across each analytical sequence. Data was filtered based upon the pooled QC samples, with compounds being retained if observed in 100 % of replicate injections for at least 1 pool and with peak area coefficient of variation (CV) <25 % across all replicate injections for each pool. A comprehensive semi-targeted approach was employed to assign the identity of urinary metabolites using an in-house compound library that included a broad range of metabolites involved in intermediary metabolism. Targeted feature extraction was performed on each dataset based on matching of metabolite chemical features against an accurate mass and retention time (AMRT) database previously generated from analysis of the IROA Technology MS metabolite library of tandards by each LC method described above, combined with the same QTOF analytical parameters used in this study (databases publicly available: https://doi.org/10.6084/m9.figshare.c.4378235.v2). In addition to accurate mass and retention time, MS/MS (data-dependent, employed for 'hit' metabolites) was also used in the confirmation of metabolite identity (i.e. level 1 identification) as per Sumner et al. Feature extraction was performed in MassHunter Profinder (build 10.0); accurate mass window of 10 ppm and retention time window of 0.3 min against the respective AMRT database. Data were exported in the format of a csv file for statistical analysis. |
| Ion Mode: | POSITIVE |
