Clinical Application of ISO and CEN/TS Standards for Liquid Biopsies-Information Everybody Wants but Nobody Wants to Pay For.
Author(s): Bonstingl L, Skofler C, Ulz C, Zinnegger M, Sallinger K, Schönberger J, Schuch K, Pankratz K, Borrás-Cherrier A, Somodi V, Abuja PM, Oberauner-Wappis L, Moser T, Heitzer E, Bauernhofer T, Kroneis T, El-Heliebi A
Publication: Clin Chem, 2024, Vol. 70, Page 1140-1150
PubMed ID: 38958115 PubMed Review Paper? No
Purpose of Paper
This paper investigated the potential effects of blood tube draw order, tube fill volume (low versus high fill level), and hemolysis status (hemolytic or non-hemolytic based on hemolysis score) on circulating cell-free DNA (ccfDNA) yield and the percentage of circulating tumor DNA (ctDNA) in plasma, as well as circulating tumor cell (CTC) status in peripheral blood specimens collected from prostate cancer patients at the onset of a change in treatment. The paper also evaluated the clinical applicability, including time requirements, of the workflows specified in recent standards for the isolation of ccfDNA from plasma and CTC staining that were released by the International Organization for Standards (ISO) and Technical Specifications from the European Committee for Standardization (CEN/TS) and sought to identify parameters associated with non-compliance.
Conclusion of Paper
Of the 659 longitudinal blood samples collected from 25 prostate cancer patients during the study, 634 (96.2%) were processed in compliance with ISO and CEN/TS documents; non-compliant samples were not inverted (1.5%) or the person collecting the specimen was not recorded (3.0%). The mean tube fill volume for all blood samples was 92.4% ± 15.8, with a fill volume ≥80% observed in 83.8% of samples. While the order of tube draw did not affect the fill volume, a significantly lower fill volume was observed for PAXgene Blood ccfDNA tubes than Acid Citrate Dextrose Solution A (ACD-A) tubes (88.8 versus 98.0%, respectively; p<0.0001).
Of the 311 samples from which plasma was isolated, 87.1% were classified as non-hemolytic (hemolysis score <0.25). The percentage of hemolytic samples did not differ significantly based on draw order (17.3%, 7.4%, and 12.2% of samples were hemolytic in tubes 1, 2, and 5, respectively; p=0.46). Hemolysis in the first tube collected (tube 1) was not predictive of hemolysis in subsequent samples (tubes 2 or 5). Hemolytic and non-hemolytic samples did not differ in in the percentage of ctDNA, ccfDNA yield, or CTC status (positive/negative), and patient-specific copy number alterations were detected in both hemolytic and non-hemolytic specimens. Hemolysis scores were comparable in CTC positive and negative samples. The percentage of ctDNA in plasma samples was weakly and positively correlated to the time that elapsed between blood draw and sample arrival in the laboratory (rs=0.2; P=0.039); no other significant correlations were observed between the percentage of ctDNA and other recorded preanalytical parameters (time elapsed between blood collection and (i) transport, (ii) initiation of CTC isolation, (iii) CTC storage), ISO and CEN/TS compliance, hemolysis score, or the duration of -80°C storage.
For ACD-A samples that underwent CTC-enrichment with the Smart Biopsy Cell Isolator, hemolysis was observed in 9.1% tubes and was weakly and negatively correlated with ISO and CEN/TS compliance (rs=-0.02, P=0.022) in these specimens; however, when the sources of noncompliance (tube inversion, identity of phlebotomist) were considered separately, the correlation was no longer significant.
For PAXgene Blood ccDNA samples, hemolysis was not significantly correlated with ISO and CEN/TS compliance, tube fill level, tube inversion, processing time, or CTC AdnaTest results, although sample sizes were small for some comparisons. However, hemolysis was weakly positively correlated with the following patient-related factors: “chemotherapy before (rs = 0.1, P = 0.036) or after blood collection (rs = 0.2, P= 0.001); the presence of lung (rs = 0.2, P = 0.003), liver (rs = 0.1, P = 0.009), or multiorgan (rs = 0.1, P = 0.019) metastases; and elevated levels of C-reactive protein (rs = 0.2, P = 0.002) or lactate dehydrogenase (LDH) (rs = 0.1, P = 0.022).”
Studies
-
Study Purpose
This study investigated the potential effects of blood tube draw order, tube fill volume (low versus high fill level), and hemolysis status (hemolytic or non-hemolytic based on hemolysis score) on circulating cell-free DNA (ccfDNA) yield, the percentage of circulating tumor DNA (ctDNA) in plasma , as well as circulating tumor cells (CTC) status in peripheral blood specimens collected from prostate cancer patients at the onset of a change in treatment. The paper also evaluated the clinical applicability, including time requirements, of the workflows specified in recent standards for the isolation of ccfDNA from plasma and CTC staining that were released by the International Organization for Standards (ISO) and Technical Specifications from the European Committee for Standardization (CEN/TS) and sought to identify parameters associated with non-compliance. Samples were considered to be ISO and CEN/TS compliant if all specified parameters and times were recorded and fell within the storage durations and temperatures specified. Blood samples (2-6 tubes per visit) were collected from 25 patients with castration-resistant prostate cancer prior to a change in therapy and at 12 week intervals thereafter (a mean of 5.4±2.8 visits per patient). The order of blood draw was fixed during the study: the first tube was not used for cfDNA analysis (tube 0), two 10 mL PAXgene Blood ccfDNA tubes (tubes 1 and 2), two 8.5 mL ACD-A tubes (tubes 3 and 4), and one additional PAXgene Blood ccfDNA tube (tube 5). The tube fill level was assessed for each sample (method not specified), and blood was transported at room temperature in a box at a constant (±2°C) temperature prior to processing and analysis (duration not reported). PAXgene Blood tubes were processed according to the ISO standard for ctDNA isolation (ISO 20186-3:2019). ccfDNA isolated from plasma (method not specified) underwent shallow sequencing on an Illumina NovaSeq machine and the ichorCNA algorithm was used to determine the percentage of ctDNA. ACD-A tubes were processed according to the CEN/TS standard for CTC isolation and staining (CEN/TS 17390-3). CTC enrichment was performed by one of two methods: the Smart Biopsy Cell Isolator (CytoGen; PAXgene Blood ccfDNA tube, tube 1) or the AdnaTest ProstateCancerPanel AR-V7 (ACD-A tubes, tubes 3 and 4). CytoGen CTC-enriched cytospins were stored at -80°C long-term (duration not reported), but AdnaTest CTC-enriched samples were processed directly (not stored). CTC status was classified as positive if any of the following markers were determined to be positive by a AdnaTest: AR-V7, AR-FL, PSA, PSMA. A sample was considered to be hemolytic if the hemolysis score, calculated by spectrophotometer, was > 0.25: hemolysis score = Absorbance at (A) 414 nm – A 385 nm + 0.1 × A 385 nm.
Summary of Findings:
Of the 659 longitudinal blood samples collected from 25 prostate cancer patients during the study, 634 (96.2%) were processed in compliance with ISO and CEN/TS documents; non-compliant samples were not inverted (1.5%) or the person collecting the specimen was not recorded (3.0%). The following represent the mean ± SD for each processing step specified, although the authors did not compare the durations of ISO and CEN/TS compliant and non-compliant samples: time from blood draw to transport = 32± 27 min, time from blood draw to arrival in the lab=45±29 min, time from blood draw to initiation of CTC isolation=62±37 min, plasma isolation process=168±71 min, CTC isolation process=248±60 min.
The mean tube fill volume for all blood samples was 92.4% ± 15.8, with a fill volume ≥80% observed in 83.8% of samples. While the order of tube draw did not affect the fill volume, a significantly lower fill volume was observed for PAXgene Blood ccfDNA tubes (Tube 5) than ACD-A tubes (Tubes 3 and 4)(88.8 versus 98.0%, respectively; P<0.0001).
Plasma was isolated from 311 blood samples, 87.1% of which were classified as non-hemolytic (hemolysis score <0.25). The percentage of hemolytic samples did not differ significantly based on tube number/ draw order (17.3%, 7.4%, and 12.2% of samples were hemolytic in tubes 1, 2, and 5, respectively; p=0.46). Hemolysis in the first tube collected (tube 1) was not predictive of hemolysis in subsequent samples (tubes 2 or 5); hemolytic samples in all tubes from a collection event only occurred 5.8% of the time, while a hemolytic sample in tube 1 and non-hemolytic samples in tubes 2 and 5 occurred in 11.6% of blood draw events.
Hemolytic and non-hemolytic samples did not differ the percentage of ctDNA, ccfDNA yield, or CTC status (positive/negative), and patient-specific copy number alterations were detected in both sample types. Hemolysis scores were comparable in CTC positive and negative samples. The percentage of ctDNA was weakly and positively correlated to the time that elapsed between blood draw and sample arrival in the laboratory (rs=0.2; P=0.039); no other significant correlations were observed between the percentage of ctDNA and the other recorded preanalytical parameters (time elapsed between blood collection and (i) transport, (ii) initiation of CTC isolation, (iii) CTC storage), ISO and CEN/TS compliance, hemolysis score, or the duration of -80°C storage.
For ACD-A samples that underwent CTC-enrichment with the Smart Biopsy Cell Isolator, hemolysis was observed in 9.1% tubes and was weakly and negatively correlated with ISO and CEN/TS compliance (rs=-0.02, P=0.022) in these specimens; however, when the sources of noncompliance (tube inversion, identity of phlebotomist) were considered separately, the correlation was no longer significant.
For PAXgene Blood ccDNA samples, hemolysis was not significantly correlated with ISO and CEN/TS compliance, tube fill level, tube inversion, processing time, or CTC AdnaTest results, although sample sizes were small for some comparisons. However, hemolysis was weakly positively correlated with the following patient-related factors: “chemotherapy before (rs = 0.1, P = 0.036) or after blood collection (rs = 0.2, P = 0.001); the presence of lung (rs = 0.2, P = 0.003), liver (rs = 0.1, P = 0.009), or multiorgan (rs = 0.1, P = 0.019) metastases; and elevated levels of C-reactive protein (rs = 0.2, P = 0.002) or lactate dehydrogenase (rs = 0.1, P = 0.022).”
Biospecimens
Preservative Types
- None (Fresh)
- Frozen
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform DNA Spectrophotometry DNA Next generation sequencing Lipid Spectrophotometry RNA RT-PCR Protein Spectrophotometry Cell count/volume Magnetic-activated cell sorting Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Aliquots and Components Hemolysis A range of hemolysis scores investigated
Biospecimen Aliquots and Components Aliquot sequential collection 1st collection
2nd collection
3rd collection
4th collection
5th collection
Biospecimen Aliquots and Components Biospecimen mixing Inverted
Non-inverted
Preaquisition Diagnosis/ patient condition Metastasis to lung, liver, or multiple organs
Elevated C-reactive protein
Elevated lactate dehydrogenase
Storage Specimen transport duration/condition Time from blood draw to arrival in the lab = 45±29
Storage Time at room temperature Time from blood draw to transport = 32± 27 min
Biospecimen Aliquots and Components Centrifugation Centrifugation delays investigated
Biospecimen Preservation Duration of tissue/ specimen processing plasma isolation process=168±71 min
CTC isolation process = 248±60 min
Preaquisition Study design ISO and CEN/TS compliant
ISO and CEN/TS non-compliant
Biospecimen Aliquots and Components Aliquot size/volume 50-100%
Preaquisition Other drugs Chemotherapy administered before blood collection
Chemotherapy administered after blood collection