A Review on the Use of Imaging Biomarkers in Oncology Clinical Trials: Quality Assurance Strategies for Technical Validation
Penny Cristinacce
Round 1
Reviewer 1 Report
Thank you for this highly interesting and comprehensive manuscript. I only have several questions/comments
1. In the table 1:
- why did you separate the PET and PET analysis analysis?
- For the relaxometry: the link "www.nist.gov/programs-projects/quantitative-mri" does'nt seem to be the correct one
- Please to homogenise the style of the police such as "Requirements on dual energy protocols & spectral imaging"
2. In Gamma Camera, it would be good to talk a little about the future direction such as ring-geometry SPECT/CT system such as Veriton or Starguide
3.Could you discuss in a separate paragraph the recent advances in AI, as it will surely contribute in all the fields you talk about.
Author Response
REVIEWER 1
Thank you for this highly interesting and comprehensive manuscript. I only have several questions/comments
1. In the table 1:
- why did you separate the PET and PET analysis analysis?
Thanks for the question. Our idea was to separate qualitative vs quantitative PET analysis but we were not enough clear. We added the word “qualitatative” to the first row. Table was moreover re-arranged to enhance its readibiility.
- For the relaxometry: the link "www.nist.gov/programs-projects/quantitative-mri" does'nt seem to be the correct one
Thanks. We checked for reference which is now correct.
- Please to homogenise the style of the police such as "Requirements on dual energy protocols & spectral imaging"
Thanks, we tried to re-formulate the table and follow your suggestion.
In Gamma Camera, it would be good to talk a little about the future direction such as ring-geometry SPECT/CT system such as Veriton or Starguide
Thanks for the idea. We added some sentence to address your suggestion.
3.Could you discuss in a separate paragraph the recent advances in AI, as it will surely contribute in all the fields you talk about.
Thanks again for the suggestion that is really appropriate. To keep the structure of the manuscript we added some sentences referring to the AI in the conclusion and discussion. Since, in general the manuscript is quite long and cover multiple topics we tried to be concise.
Reviewer 2 Report
This review from Chauvie et al focuses on a very delicate issue regarding intercomparison of quantitative image-based outcomes in clinical trials among different centers. Even though it is of peculiar interest for medical physics expert (MPEs), the aspects discussed are pertinent to the workflow of other specialists involved in the diagnostic imaging chain, such as radiologists, nuclear medicine specialists, radiotherapists, as well as technologists. Some quality assurance aspects are not specific for oncological patients, as the title suggest, but are anyway helpful to more comprehensively address the problem of imaging biomarker reproducibility.
I just have a few recommendations:
1) Optimization strategies for PET (Section 2.1.4). The use of phantoms with hollow sphere is standard practice, but it has the known pitfall related to the thickness of the spherical cavities, separating the tumor from the background. Investigators are developing 3D printed phantoms to overcome this limitation, even though this is far from becoming standard practice in the near future. 3D printing has also the advantage of reproducing irregular shapes, instead of just sphere, obtaining more reproducible partial volume related inaccuracies in the uptake measurement. I would suggest the authors to cite some papers proposing this strategy, such as
- Gillett D, Marsden D, Ballout S, Attili B, Bird N, Heard S, Gurnell M, Mendichovszky IA, Aloj L. 3D printing 18F radioactive phantoms for PET imaging. EJNMMI Phys. 2021 Apr 28;8(1):38. doi: 10.1186/s40658-021-00383-6. PMID: 33909154; PMCID: PMC8081805.
- F. Gallivanone et al 2016 JINST 11 C01022, DOI 10.1088/1748-0221/11/01/C01022
2) Due to the increasing role of radiomics and radiogenomics in oncology, a specific subsection would be required. The European Society of Radiology has recently published an expert consensus on this topic, which is fully within the scope of this new manuscript:
- Fournier, L., Costaridou, L., Bidaut, L. et al. Incorporating radiomics into clinical trials: expert consensus endorsed by the European Society of Radiology on considerations for data-driven compared to biologically driven quantitative biomarkers. Eur Radiol 31, 6001–6012 (2021). https://doi.org/10.1007/s00330-020-07598-8
Also, EANM and SNMMI have jointly published guidelines regaring radiomics in nuclear medicine, covering aspects related to harmonization and standardization:
- Hatt, M., Krizsan, A.K., Rahmim, A. et al. Joint EANM/SNMMI guideline on radiomics in nuclear medicine. Eur J Nucl Med Mol Imaging 50, 352–375 (2023). https://doi.org/10.1007/s00259-022-06001-6
The Imaging Biomarkers Standardization Initiative (IBSI) must also be mentioned in this context:
- Zwanenburg A, Vallières M, Abdalah MA, et al. The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping. Radiology. 2020 May;295(2):328-338. doi: 10.1148/radiol.2020191145. Epub 2020 Mar 10. PMID: 32154773; PMCID: PMC7193906.
The European Imaging Biomarkers Alliance (EIBALL - merging previous activities ESR Subcommittee on Imaging Biomarkers and other ESR working groups) deserves a mention along with the already mentioned QIBA - RSNA.
3) Section 2.4 on CT is missing a subsection specific for the newly introduced photon counting CT, having the potential to overcome many issues related to the HU reproducibility. Cite, for instance:
- Liu, L.P., Shapira, N., Chen, A.A. et al. First-generation clinical dual-source photon-counting CT: ultra-low-dose quantitative spectral imaging. Eur Radiol 32, 8579–8587 (2022). https://doi.org/10.1007/s00330-022-08933-x
- Meloni, A.; Frijia, F.; Panetta, D.; Degiorgi, G.; De Gori, C.; Maffei, E.; Clemente, A.; Positano, V.; Cademartiri, F. Photon-Counting Computed Tomography (PCCT): Technical Background and Cardio-Vascular Applications. Diagnostics 2023, 13, 645. https://doi.org/10.3390/diagnostics13040645
4) Table I must be checked for uniformity of font and text style
Author Response
REVIEWER 2:
This review from Chauvie et al focuses on a very delicate issue regarding intercomparison of quantitative image-based outcomes in clinical trials among different centers. Even though it is of peculiar interest for medical physics expert (MPEs), the aspects discussed are pertinent to the workflow of other specialists involved in the diagnostic imaging chain, such as radiologists, nuclear medicine specialists, radiotherapists, as well as technologists. Some quality assurance aspects are not specific for oncological patients, as the title suggest, but are anyway helpful to more comprehensively address the problem of imaging biomarker reproducibility.
I just have a few recommendations:
1) Optimization strategies for PET (Section 2.1.4). The use of phantoms with hollow sphere is standard practice, but it has the known pitfall related to the thickness of the spherical cavities, separating the tumor from the background. Investigators are developing 3D printed phantoms to overcome this limitation, even though this is far from becoming standard practice in the near future. 3D printing has also the advantage of reproducing irregular shapes, instead of just sphere, obtaining more reproducible partial volume related inaccuracies in the uptake measurement. I would suggest the authors to cite some papers proposing this strategy, such as
- Gillett D, Marsden D, Ballout S, Attili B, Bird N, Heard S, Gurnell M, Mendichovszky IA, Aloj L. 3D printing 18F radioactive phantoms for PET imaging. EJNMMI Phys. 2021 Apr 28;8(1):38. doi: 10.1186/s40658-021-00383-6. PMID: 33909154; PMCID: PMC8081805.
- F. Gallivanone et al 2016 JINST 11 C01022, DOI 10.1088/1748-0221/11/01/C01022
We thank the reviewer for pointing these out. We agree that 3D printing and even gelatin and PVA cryogel phantoms (both of which have been investigated for >10 years) present many advantages, albeit likely limited to research institutions with sufficient resources. However their inclusion is warranted, and we have added a general statement about them to section 2.2, and the following references. We hope this satisfies the comment.
- Gillett D et al. 3D printing 18F radioactive phantoms for PET imaging. EJNMMI Phys. 2021 Apr 28;8(1):38. doi: 10.1186/s40658-021-00383-6.
Soultanidis GM et al. PVA cryogel for construction of deformable PET-MR visible phantoms. IEEE Trans Nucl Sci. 2013;60(1):95–102. doi: 10.1109/TNS.2013.2238952.
2) Due to the increasing role of radiomics and radiogenomics in oncology, a specific subsection would be required. The European Society of Radiology has recently published an expert consensus on this topic, which is fully within the scope of this new manuscript:
- Fournier, L., Costaridou, L., Bidaut, L. et al. Incorporating radiomics into clinical trials: expert consensus endorsed by the European Society of Radiology on considerations for data-driven compared to biologically driven quantitative biomarkers. Eur Radiol 31, 6001–6012 (2021). https://doi.org/10.1007/s00330-020-07598-8
Also, EANM and SNMMI have jointly published guidelines regaring radiomics in nuclear medicine, covering aspects related to harmonization and standardization:
- Hatt, M., Krizsan, A.K., Rahmim, A. et al. Joint EANM/SNMMI guideline on radiomics in nuclear medicine. Eur J Nucl Med Mol Imaging 50, 352–375 (2023). https://doi.org/10.1007/s00259-022-06001-6
The Imaging Biomarkers Standardization Initiative (IBSI) must also be mentioned in this context:
- Zwanenburg A, Vallières M, Abdalah MA, et al. The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping. Radiology. 2020 May;295(2):328-338. doi: 10.1148/radiol.2020191145. Epub 2020 Mar 10. PMID: 32154773; PMCID: PMC7193906.
The European Imaging Biomarkers Alliance (EIBALL - merging previous activities ESR Subcommittee on Imaging Biomarkers and other ESR working groups) deserves a mention along with the already mentioned QIBA - RSNA.
We thank the reviewer for this nice summary of the professional guidance on Radiomics. We had initially purposefully omitted Radiomics from the manuscript but in hindsight, the high amount of work using Radiomic features as IBs demonstrates that it is well deserved to mention it. Rather than add a specific section on Radiomics, we wrote 2 dedicated paragraphs discussing multimodality radiomics endeavors, and opted to add it to the Discussion. We think it “fits” better discussed here and hope this satisfies the comment. We also added a number of new references to support the text.
3) Section 2.4 on CT is missing a subsection specific for the newly introduced photon counting CT, having the potential to overcome many issues related to the HU reproducibility. Cite, for instance:
- Liu, L.P., Shapira, N., Chen, A.A. et al. First-generation clinical dual-source photon-counting CT: ultra-low-dose quantitative spectral imaging. Eur Radiol 32, 8579–8587 (2022). https://doi.org/10.1007/s00330-022-08933-x
- Meloni, A.; Frijia, F.; Panetta, D.; Degiorgi, G.; De Gori, C.; Maffei, E.; Clemente, A.; Positano, V.; Cademartiri, F. Photon-Counting Computed Tomography (PCCT): Technical Background and Cardio-Vascular Applications. Diagnostics 2023, 13, 645. https://doi.org/10.3390/diagnostics13040645
We thank the reviewer for referencing this novel emerging technology. We do already discuss PCCT in section 2.4.2 and the added IBs that are brought, but more specifically in relation to the more reproducible HU that are brought by PCCT. We made a few changes to section 2.4.2 to make this more explicit
4) Table I must be checked for uniformity of font and text style
We have made changes to Table 1 based on comments from other reviewers, and have also checked the Table for style consistency
Reviewer 3 Report
The aim of this manuscript is to review a range of quantitative imaging biomarkers used in oncology clinical trials and give an overview of the quality assurance considerations from the point of view of the medical physics expert. It highlights the need to consider the entire imaging chain, from data acquisition through to analysis and, as such, it could be a welcome addition to the literature. However, I believe the manuscript needs some significant improvement to deliver the key messages more effectively.
The abstract, introduction and discussion are, in general, well written and give a good overview of the aims and objectives of the work. A few points of note from the Introduction:
L16 …precise, accurate and reproducible.
L50 Give references for some of the biomarkers that guide clinical decision making. Eg deSouza et al https://doi.org/10.1186/s13244-019-0764-0
L52 This is a strange place to cite ref [18], the only roadmap to clinical translation of quantitative imaging biomarkers we have. Suggest move to L165.
L82 Use MPEs consistently. MPs used in places.
L95 Remove supplementary from Supplementary Table 1 or make table supplementary.
The choice of qIBs described in the manuscript is justified in section 2. Whilst I believe that they are imaging biomarkers of interest, the reasoning for their choice is unclear. I think it would be better to reference to the deSouza et al paper linked above. The authors are also advised to look through EIBALL's Biomarker Inventory https://www.myesr.org/research/biomarkers-inventory to ensure that the choice of biomarkers is built upon previously published works and ideally not just “the authors’ experience”.
The manuscript relies quite heavily on the important work of QIBA, the author’s should extend the text to other international bodies looking at these issues, such as EIBALL, QIN, NCITA etc.
The following reference (and the references therein) are recommended:
Delivering a Quantitative Imaging Agenda. deSouza et al. Cancers 2023, 15(17), 4219 https://doi.org/10.3390/cancers15174219
Challenges in ensuring the generalizability of image quantitation methods for MRI. Keenan et al. Medical Physics. 49(4), 2820. https://doi.org/10.1002/mp.15195
Clinical translation of quantitative magnetic resonance imaging biomarkers – An overview and gap analysis of current practice. Hubbard Cristinacce et al. Physica Medica 2022, 101, 165. https://doi.org/10.1016/j.ejmp.2022.08.015
Introduction to the National Cancer Imaging Translational Accelerator (NCITA): a UK-wide infrastructure for multicentre clinical translation of cancer imaging biomarkers. McAteer et al. BJC 2021. 125, 1462. https://doi.org/10.1038/s41416-021-01497-5
QIN Benchmarks for Clinical Translation of Quantitative Imaging Tools. Farahani et al. Tomography 2019. 5(1) 1-6. https://doi.org/10.18383/j.tom.2018.00045
ESR Statement on the Validation of Imaging Biomarkers. Alberich-Bayarri et al. Insights into Imaging 2020. 11, 76. https://doi.org/10.1186/s13244-020-00872-9
Table The table could be much improved. The layout is poor and confusing (although I appreciate that this is probably no fault of the authors). Its difficult to differentiate the entries from the different rows, the formatting is erratic, and the centred text makes it hard to read. The entries need more description of what is meant by the headings: QA testing, harmonization, dosimetry and advanced analysis. I would suggest this is either done in the relevant text or in the table caption. The order of the IBs in the table does not match that in the text. The order of IBs in both the text and the table isn’t clear – it could be improved by going from the most to least standardised perhaps? This could help the narrative in the text. Overall I think it needs more input and refinement from subject matter experts in each field and more consistency in the length of the entries. The authors could consider adding references to allow the reader to find the necessary detail about each bullet point. A few extra points:
1) Dynamic contrast enhanced, not direct contrast enhanced
2) QIBA, not QUIBA
3) Do not say ‘Same as…” as the reader not have read the previous entry.
4) Full stops need to be consistent.
Other comments from page 8 onwards
L41 “however as of yet, MTV and TGV do not have the three qualities of precision, accuracy and repeatability” Consider changing to “however as of yet, the precision, accuracy and repeatability of MTV and TGV have not been established.”
L50 “has, to date, limited application in routine clinical practice”.
L58 “and on the patient themselves.” Reword. Currently sounds like there’s an error with the patient.
L104 Give references.
L185 “The definition and subsequent translation into the clinic…”
L187 “…many strategies that allow us to estimate parameters…”
L207 Give references.
L218 Suggest removing “disturbing”
L244 Ktrans, ve, and Kep. Amend superscript/subscripts.
L270 Link to QIN page doesn’t work. Need to be clearer about what QIN have done/doing. See references above.
L275 Extra full stop.
L307 “..a control image and an image tagged via an additional selective radio frequency pulse…”
L314 “is the most commonly used IB in ASL.”
L336 The text summarising ADC uses the term b-value, which is not explained and also states that only two b-values are used, when it should say at least two. Reword. Link to Barnes et al. https://doi.org/10.1259/bjr.20170577 somewhere in DWI section.
L388 Mention and reference NIST-ISMRM diffusion phantom.
Reviews such as this are a welcome addition to the literature and important for understanding the QA needs of quantitative IBs and their clinical translation.
The English could be improved in place, some of which I detail in my comments. The introduction and discussion read well, but there are inconsistencies in the style between the PET, MRI and CT sections. I think the MR section requires the most attention, but I may be more sensitive to this as it’s my area of expertise.
Author Response
POINT BY POINT RESPONSE TO REVIEWER 3
The aim of this manuscript is to review a range of quantitative imaging biomarkers used in oncology clinical trials and give an overview of the quality assurance considerations from the point of view of the medical physics expert. It highlights the need to consider the entire imaging chain, from data acquisition through to analysis and, as such, it could be a welcome addition to the literature. However, I believe the manuscript needs some significant improvement to deliver the key messages more effectively.
The abstract, introduction and discussion are, in general, well written and give a good overview of the aims and objectives of the work. A few points of note from the Introduction:
L16 …precise, accurate and reproducible. Done
L50 Give references for some of the biomarkers that guide clinical decision making. Eg deSouza et al https://doi.org/10.1186/s13244-019-0764-0. Done
L52 This is a strange place to cite ref [18], the only roadmap to clinical translation of quantitative imaging biomarkers we have. Suggest move to L165. Thank you for the suggestion: we moved the reference where requested by the reviewer.
L82 Use MPEs consistently. MPs used in places. Thank you for the suggestion: corrections have been done through the text
L95 Remove supplementary from Supplementary Table 1 or make table supplementary. Done
The choice of qIBs described in the manuscript is justified in section 2. Whilst I believe that they are imaging biomarkers of interest, the reasoning for their choice is unclear. I think it would be better to reference to the deSouza et al paper linked above. The authors are also advised to look through EIBALL's Biomarker Inventory https://www.myesr.org/research/biomarkers-inventory to ensure that the choice of biomarkers is built upon previously published works and ideally not just “the authors’ experience”.
The manuscript relies quite heavily on the important work of QIBA, the author’s should extend the text to other international bodies looking at these issues, such as EIBALL, QIN, NCITA etc.
The following reference (and the references therein) are recommended:
Delivering a Quantitative Imaging Agenda. deSouza et al. Cancers 2023, 15(17), 4219 https://doi.org/10.3390/cancers15174219
Challenges in ensuring the generalizability of image quantitation methods for MRI. Keenan et al. Medical Physics. 49(4), 2820. https://doi.org/10.1002/mp.15195
Clinical translation of quantitative magnetic resonance imaging biomarkers – An overview and gap analysis of current practice. Hubbard Cristinacce et al. Physica Medica 2022, 101, 165. https://doi.org/10.1016/j.ejmp.2022.08.015
Introduction to the National Cancer Imaging Translational Accelerator (NCITA): a UK-wide infrastructure for multicentre clinical translation of cancer imaging biomarkers. McAteer et al. BJC 2021. 125, 1462. https://doi.org/10.1038/s41416-021-01497-5
QIN Benchmarks for Clinical Translation of Quantitative Imaging Tools. Farahani et al. Tomography 2019. 5(1) 1-6. https://doi.org/10.18383/j.tom.2018.00045
ESR Statement on the Validation of Imaging Biomarkers. Alberich-Bayarri et al. Insights into Imaging 2020. 11, 76. https://doi.org/10.1186/s13244-020-00872-9
We thank the reviewer for this important comment, which we believe will make the text more complete. Initiatives promoted by international bodies other than the RSNA QIB have been addressed in the text and suggested references have been added. The text has been modified accordingly.
Table The table could be much improved. The layout is poor and confusing (although I appreciate that this is probably no fault of the authors). Its difficult to differentiate the entries from the different rows, the formatting is erratic, and the centred text makes it hard to read. The entries need more description of what is meant by the headings: QA testing, harmonization, dosimetry and advanced analysis. I would suggest this is either done in the relevant text or in the table caption. The order of the IBs in the table does not match that in the text. The order of IBs in both the text and the table isn’t clear – it could be improved by going from the most to least standardised perhaps? This could help the narrative in the text. Overall I think it needs more input and refinement from subject matter experts in each field and more consistency in the length of the entries. The authors could consider adding references to allow the reader to find the necessary detail about each bullet point. A few extra points:
1) Dynamic contrast enhanced, not direct contrast enhanced
2) QIBA, not QUIBA
3) Do not say ‘Same as…” as the reader not have read the previous entry.
4) Full stops need to be consistent.
We thank the reviewer for the suggestion regarding the table, which was indeed difficult to read.
The table has been modified in both layout and content. The amount of information has been reduced. The list of biomarkers has been defined following the type of system (starting with PET, then SPECT, etc.). Since this was not clear in the previous version, the table header was changed to describe this. In general, the table headings have been expanded to emphasize how Table 1 is a guide with suggestions for practical implementation of the QIBs provided in the text. Extra points have been corrected as requested.
Other comments from page 8 onwards
Thanks to the reviewer for all comments below, all of which were addressed in the text review.
L41 “however as of yet, MTV and TGV do not have the three qualities of precision, accuracy and repeatability” Consider changing to “however as of yet, the precision, accuracy and repeatability of MTV and TGV have not been established.”
L50 “has, to date, limited application in routine clinical practice”.
L58 “and on the patient themselves.” Reword. Currently sounds like there’s an error with the patient.
L104 Give references.
L185 “The definition and subsequent translation into the clinic…”
L187 “…many strategies that allow us to estimate parameters…”
L207 Give references.
L218 Suggest removing “disturbing”
L244 Ktrans, ve, and Kep. Amend superscript/subscripts.
L270 Link to QIN page doesn’t work. Need to be clearer about what QIN have done/doing. See references above.
L275 Extra full stop.
L307 “..a control image and an image tagged via an additional selective radio frequency pulse…”
L314 “is the most commonly used IB in ASL.”
L336 The text summarising ADC uses the term b-value, which is not explained and also states that only two b-values are used, when it should say at least two. Reword. Link to Barnes et al. https://doi.org/10.1259/bjr.20170577 somewhere in DWI section.
L388 Mention and reference NIST-ISMRM diffusion phantom.
Round 2
Reviewer 2 Report
I'm satisfied with the author's amendments.
Author Response
We again thank the reviewer for the comments that helped in improving the manuscript
Reviewer 3 Report
I'd like to thank the authors for including my suggestions in the updated manuscript. The text is now more comprehensive and easier to read. However, there are still a few minor changes that would benefit the manuscript.
52 - reference both 18 and 19 here (O'Connor and de Souza)
67 - reference 18 (O'Connor)
Table 1 - line 110 "starting with PET"
- Pre-trial quality
- remove reference to Gd contrast in DWI patient safety section
145 - cm3
245 - "an analogue", not "an analogous"?
248 - "in a few years" not "from a few years"
249 - CZT needs to be written in full
347 - "patient movement" not "patient’s slight movement"
412 - Arterial input function has been abbreviated on line 391.
430 - Add some reference to the work of OSIPI, who have been doing a lot of work in the area of quantitative perfusion MR standardisation (ample references: https://osipi.github.io/; https://pubmed.ncbi.nlm.nih.gov/37279059/; https://doi.org/10.1002/mrm.29826; https://doi.org/10.1002/mrm.29840 (early view in the next week) etc).
436 - "positioned on a feeding artery for the volume of interest" not "positioned below the volume of interest"
446 - Remove "The ease of use and"
527 - Remove gap in T2*
720 - Give references to "good reviews are available"
729 - acceptance testing?
There are quite a few typographical errors throughout that need to be addressed by the copyeditor.
The English is acceptable.
Author Response
We really want to thanks the reviewer for the careful and comprehensive reading of the manuscript: all suggested changes (see below) have been made.
52 - reference both 18 and 19 here (O'Connor and de Souza) ok
67 - reference 18 (O'Connor) ok
Table 1 - line 110 "starting with PET" ok
- Pre-trial quality ok
- remove reference to Gd contrast in DWI patient safety section ok
145 - cm3 ok
245 - "an analogue", not "an analogous"? ok
248 - "in a few years" not "from a few years" ok
249 - CZT needs to be written in full ok
347 - "patient movement" not "patient’s slight movement" ok
412 - Arterial input function has been abbreviated on line 391 ok
430 - Add some reference to the work of OSIPI, who have been doing a lot of work in the area of quantitative perfusion MR standardisation (ample references: https://osipi.github.io/; https://pubmed.ncbi.nlm.nih.gov/37279059/; https://doi.org/10.1002/mrm.29826; https://doi.org/10.1002/mrm.29840 (early view in the next week) etc).
Thank you for this important suggestion. References to the OSIPI works have been added in the Discussion and Conclusions section.
436 - "positioned on a feeding artery for the volume of interest" not "positioned below the volume of interest" ok
446 - Remove "The ease of use and" ok
527 - Remove gap in T2* ok
720 - Give references to "good reviews are available"
References 206-211 have been added
729 - acceptance testing? ok
