Frequently Asked Questions

Version 2.5 26.03.2018
Help us to improve and further develop this frequently asked questions! Send your question to support@pirche.com.

What means PIRCHE?

The Prediction of Indirectly ReCognized HLA Epitopes (PIRCHE) is a new algorithm to estimate the potential of a T-cell related immune response after transplantation. In hematopoietic stem cell transplantation (HSCT), PIRCHE predicts graft versus host immune response against mismatched HLA. In solid organ transplantation (SOT), PIRCHE predicts de novo development of donor specific antibodies.

The PIRCHE algorithm has the goal to provide additional data to support the decision about the best donor for a particular patient. Eric Spierings, PhD is the inventor of the PIRCHE algorithm (for details see the list of publications) and his working group is constantly improving the algorithm.

Indirect T-cell recognition of allogeneic HLA depends on HLA derived peptides that differ between host and graft. These HLA-derived peptides are likely presented on shared HLA. HLA-derived peptides being identical between donor and recipient should be ignored by the alloimmune system, as T-cells recognizing these peptides should have been deleted from the repertoire due to thymic selection. Thus, the GvHD reaction after HSCT should be evoked by recipient-specific peptides. In order to predict permissibility of HLA mismatches PIRCHE in silico predicts the presentation of allogeneic HLA derived peptides on shared HLA. Any difference in presentable peptides derived from donor-versus recipient alleles is regarded a PIRCHE.

Retrospective studies showed that HLA mismatched HSCT with high numbers of both, PIRCHE I and PIRCHE II are correlated to clinical alloreactivity (see references).

What is the function of the PIRCHE-Module?

The PIRCHE-Module is a simple web browser application to calculate PIRCHE scores.

Scores shown in the PIRCHE I column of the PIRCHE Module represent the sum of HLA derived peptides predicted to be presented by shared HLA class I molecules. The PIRCHE II column lists the sum of HLA derived peptides that are predicted to be presented by shared HLA class II molecules.

The PIRCHE-Module is for research use only.

It demonstrates the functionality of the PIRCHE algorithm providing an impression of the benefit for existing patient-donor matching systems, if integrated. Let us know if you want to learn more about how to integrate the PIRCHE algorithm into your patient-donor matching system.

What HLA typing is required for PIRCHE analysis?

HLA data of the patient and the potential donors need to be typed unambiguously on HLA allele level.

How to start a PIRCHE analysis?

See our tutorial on how to use the PIRCHE-Module.

What does “Single Patient” vs. “Multi Patient (CSV)” mean?

PIRCHE-Module offers you 2 ways to enter your data and to receive the results. If you want to analyze the PIRCHE scores of a single patient, the “Single Patient” mode works best for you. Here you can enter data of a single patient and a list of potential donors.

If you want to analyze PIRCHE scores for a list of patients with their connected potential donors, you may use the “Multi Patient (CSV)” mode. Here you can provide all your data in a single block and you will receive a single CSV-text output.

See our tutorial for further information on how to input data.

How to enter patient and donor data in “Single Patient” mode?

The “Single Patient” mode is the standard way to enter patient and donor data. Patient data are entered as a simple comma separated text string into the patient ID, HLA data field. Start patient data with the patient ID. Any anonymous and unique combination of alphanumeric and special (-_!“§$%&/()=?#+*) characters are allowed. It is recommended to use patient IDs which are shorter than 30 characters. The patient ID is concluded by “,” or “;”. Enter patient HLA data consecutively after the patient ID in allele specific HLA format in the order of your choice (see example in Fig 1. and http://hla.alleles.org/nomenclature/naming.html). HLA alleles shall be separate by “,” or “;”. In case of homozygous loci, you may provide only one or both alleles (see example don-036 and don-037 in table 1).

Enter your donor data in the same way into the Donor List ID, HLA data field starting a new line for each of the potential donors.

See also our tutorial to learn about entering data.

Table 1. Example of data entry format for “Single Patient” mode

Patient ID, HLA data

Patient-4711, A*24:02, A*68:02, B*44:03, B*53:01, C*04:01, C*01:02, DRB1*07:01, DRB1*15:01, DQB1*02:02, DQB1*06:02

Donor List ID, HLA data

don-095, A*23:01, A*68:02, B*44:03, B*53:01, C*04:01, C*01:02, DRB1*07:01, DRB1*15:01, DQB1*02:02, DQB1*06:02
don-042, A*24:03, A*68:02, B*44:03, B*53:01, C*04:01, C*01:02, DRB1*07:01, DRB1*15:01, DQB1*02:02, DQB1*06:02
don-036, A*24:02, A*68:02, B*44:03, B*48:01, C*01:02, C*01:02, DRB1*07:01, DRB1*15:01, DQB1*02:02, DQB1*06:02
don-037, A*24:02, A*68:02, B*44:03, B*48:01, C*01:02, DRB1*07:01, DRB1*15:01, DQB1*02:02, DQB1*06:02

How to enter patient and donor data in “Multi Patient (CSV)” mode?

The “Multi Patient (CSV)” mode is made to enter multiple combinations of patient and donor data. Data are entered as simple text into the Patient / Donor List ID, HLA data field.

Patient and donor data follow the same format description outlined for the “Single Patient” mode. In the “Multi Patient (CSV)” mode a set of patient and connected donor data are provided in consecutive lines. The next set of patient and donor data is separated by a line only carrying a “,” (see example in table 2).

 

Table 2. Example of data entry format for “Multi Patient (CSV)” mode

Patient / Donor List ID, HLA data

Pat-7411,A*24:02,A*68:02,B*44:03,B*53:01,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
don-996,A*24:02,A*68:02,B*44:03,B*53:01,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
don-095,A*23:01,A*68:02,B*44:03,B*53:01,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
,
Pat-7412,A*23:02,B*44:03,B*53:01,C*05:01,C*01:02,DRB1*10:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
don-036,A*24:02,A*68:02,B*44:03,B*48:01,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
don-311,A*24:02,A*68:02,B*44:03,B*53:01,C*04:01,C*01:02,DRB1*07:01,DRB1*01:01,DQB1*02:02,DQB1*06:02
don-530,A*24:02,A*68:02,B*44:03,B*53:01,C*02:02,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
,
Pat-7413,A*24:02,B*45:03,B*51:01,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*08:02
don-188,A*24:02,A*68:02,B*44:03,B*44:03,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*02:02,DQB1*06:02
don-782,A*24:02,A*68:02,B*44:03,B*53:01,C*04:01,C*01:02,DRB1*07:01,DRB1*15:01,DQB1*03:03,DQB1*06:02

What is a “Search Profile”?

The “Search Profile” mode allows to find suitable donors with only one allelic mismatch to the given patient. By carefully filtering the list of suggested donors (by PIRCHE score, mismatch and frequency), the user obtains a set of eligible donors. Scanning the search registries for these donors might be a simplified task compared to searching with the given patient typing.

How to handle the result of a PIRCHE analysis?

In the “Single Patient” mode you may sort your result by clicking on the PIRCHE I or PIRCHE II column header. A little arrow indicates the sorting direction.

Clicking on a locus’ column header filters the table for donors that have no mismatch on the selected locus.

In the “Filter donor ID” field you may select for particular donors by entering parts of the donor ID. The “X“ deletes any filtering.

The “Back” button brings back the patient and donor data view.

To save the result you may use the standard browser functions “select all”, “copy”, “paste” or printing to a PDF. The table can easily be pasted into standard software solutions for further analysis.

In case of the “Multi Patient (CSV)” mode results of the PIRCHE analysis are provided in CSV-text output. The results can easily be transferred by standard browser functions “select all”, “copy” and “paste” into other software tools (e.g. spreadsheet tools) and then subsequently saved as csv-file.

See our tutorial to learn how to read PIRCHE results.

Can I see details of the PIRCHE analysis?

By clicking on a result line in the “Single Patient” mode, the details view opens up. From left to right following details are shown: HLA ID (link to presenting shared HLA allele in IMGT/HLA database), Allele Code (HLA name of presenting shared HLA molecule), Core Sequence (predicted presented peptide core sequence of the patient at mismatched HLA locus), Peptide (in case of HLA class II), IC 50 Score (predicted IC50 binding score).

Details are also provided in a single line format in the “Multi Patient (CSV)” mode. The header lists the sequence of details provided.

See our tutorial to learn more about the details.

What are known limitations of the PIRCHE algorithm?

The PIRCHE-Module is made for the PIRCHE score calculation to support the selection of donors or cord blood units for hematopoietic stem cell transplantation.

The PIRCHE-Module is for research use only. The use of PIRCHE AGs PIRCHE-Module does not provide any advice on the usability of donor or stem cells, nor does PIRCHE AG check the patient and donor data provided by the user for accuracy, completeness, legality, correctness or actuality.

Studies performed (see list of publications) analyzed the PIRCHE algorithm in cases of single or rare mismatches (e.g. 09/10). You should not use the PIRCHE algorithm for the analysis of multiple mismatches between patients and donors.

For every HLA locus reported for the patient the same HLA locus needs to be provided for all donors.

Will my patient and donor data be stored on the server?

For accounting purposes only the patient ID in clear text and the patient ID together with the associated HLA pattern in encrypted form (salted SHA-256 hash-value) will be stored. To ensure privacy, you should provide generic patient IDs (e.g. “1234”).

No donor data is stored on the server.

After processing your PIRCHE request, PIRCHE AG has no access to any typing information.

How many patients/donors can be matched at once?

The “Single Patient” mode stays responsive even when matching several hundred donors.

The “Multi Patient (CSV)” mode is capable of handling thousands of matching operations at once.

The “Search profile” mode only handles one patient at a time.

How can I interpret the PIRCHE scores?

PIRCHE is a new algorithm to estimate the potential of a T-cell related graft versus host immune response against mismatching HLA after transplantation. The goal of the PIRCHE algorithm is to provide additional data to support the decision making process about the best donor for the particular patient, in case no full matched donor is available.

Published studies indicate, that patient-donor pairs with low PIRCHE I and II show better follow up parameter than patient-donor pairs with high PIRCHE I and II values, including better overall survival (OS), better disease-free survival, transplant-related mortality (TRM) and better acute and chronic graft-versus-host disease (GvHD).