IT systems · Practical guide

Validating Agilent MassHunter: tracing the result back to the LC/MS run

An Agilent MassHunter analysis upgrade leaves the acquisition component on its previous release. An old result opens, but reprocessing changes the value. Assess the component combination, investigate the difference and establish independent expected results for the LC/MS workflow. Trace acquisition and evaluation as separate parts of the data path.

Table of contents

At a glance

How do you validate Agilent MassHunter for GxP use?

Validate the actual combination of instrument, acquisition, processing, storage and user administration for its intended use. Trace measurements, method versions, processing and results together. Assess the relevance of supplier evidence. Opening a historical result, reprocessing it and changing it manually are separate test cases. The MassHunter family name or a single version number is not a sufficient system description.

  • Data capture and evaluation have separate versions and dependencies.
  • The OpenLab-/ECM integration is determined by the specific operational setup.
  • The results display and repeated calculation require different target criteria.
  • Data format, method reference and editing history belong in the proof.

Which MassHunter components are part of your system?

Agilent describes MassHunter Acquisition for LC/MS as instrument control and data collection for various LC/MS device families. Qualitative analysis and quantitative analysis serve distinct functions. Therefore, start with the actual analytical pathway—both measurement and data evaluation—used in practice. Simply stating "MassHunter is installed" leaves critical dependencies unaddressed.

The current document status is intentionally inconsistent: Qualitative Analysis 14.0 is described in Release Notes from May 2026. The separately evaluated LC/TQ Acquisition Release Notes carry version 12.3 and date May 2025. These represent component-specific evidences, not a consolidated released combination, and do not indicate the status of any latest patch.

Capture instrument, control software, evaluation programs, operating system, data storage and connected services with their real levels. Arrange a proof for each transition: Who generates the record, who changes it, where it is stored and what later reads it? GC/MS or ICP-MS functions only take over from the appropriate documents.

Sources: [1] [2] [3]

Further reading:

Define the analytical purpose prior to the test list.

This guide uses the example of an LC/MS application in the quality control of human medicinal products. For computerized GMP activities, Annex 11 of the EU GMP guidelines serves as a key reference. The current EudraLex index still references the January 2011 revision. Other laboratory purposes require separate classification.

Describe specifically which decision is supported by a result. An identity test, a quantitative content assessment, and an exploratory evaluation entail different requirements. Document who commissions measurements, specifies methods, evaluates results, and approves usage. Thus, "ensure data integrity" becomes a verifiable requirement.

System validation does not automatically establish the analytical suitability of a method. Conversely, an appropriate method does not substitute for evidence of data processing. For FDA-related records, you must additionally clarify the actual Part 11 scope based on the underlying recordkeeping obligations. The product name neither creates a universal FDA requirement nor constitutes approval.

Sources: [4] [5] [6] [7]

Use a component matrix to avoid assumptions about versions

The following work aid consolidates various levels. For your project, enter the installed build and the corresponding manufacturer documentation. A manufacturer's record is acceptable if purpose, version, configuration, and expected behavior align. An additional review closes the remaining specific gap.

Particularly important is the version matching between Acquisition and Evaluation. The older LC/TQ-12.3 documentation lists Qualitative 13.0 as a supported version. Qualitative 14.0 specifies different, more recent dependencies. Therefore, compatibility cannot be substantiated from two separately available programs.

Own work aid: Components and required assignment
ComponentSupported referenceTo demonstrate the project
LC/TQ-Entry Acquisition 12.3, Document May 2025 Device, Build, Driver, Worklist, and generated dataset
Qualitative Evaluation Qualitative Analysis 14.0, May 2026 Appropriate input data, current methods, and intended processing
Quantitative Evaluation Independent component; listed here only as a dependency Valid instruction for used integration, calculation, and result statuses
Data retention and logging Configuration-dependent local or central services Actual storage location, authentication, and active authorizations

Sources: [2] [3] [5]

Consider both data format and operating mode together

Qualitative Analysis 14.0 describes the Workstation, Networked Workstation, and an OpenLab-CDS Client/Server context. This results in OpenLab integration being occupied in certain configurations. It is neither required nor generally excluded in every MassHunter installation. The actual architecture determines the memory and management limits.

The version status supports .d and .sirslt; for .sirslt, a sample audit trail is described. An export to .d is also documented. This does not imply identical information coverage between the formats. When performing an export, verify which data and relationships are transferred and which information remains required in the source.

A useful acceptance question is: Can an authorized reviewer understand the selected result including its origin without relying on verbal explanations? Establish a representative dataset for this purpose. Compare the original data, analysis, export, and if applicable, the report against the actually required information.

Sources: [2] [5] [6]

Further reading:

Open, reprocess, and manually modify differ

Release notes for Qualitative 14.0 distinguish between the unchanged display of previous results and results from re-executed workflows or manual actions. Altered algorithms may produce different outcomes upon reprocessing. Therefore, a version change requires separate expectations for pure display and for new calculations.

Define prior to comparison which input data and method conditions remain unchanged. Document any deliberately modified parameter. Otherwise, a difference cannot be meaningfully attributed to a specific algorithm, method, or data set. A discrepant numerical value is initially considered an investigation-worthy finding.

Review the user action, authorization, and history of a manual integration within the actually used analysis program. For the review, determine which original and subsequent results are required. Neither every change nor every repetition automatically constitutes an out-of-specification (OOS) result or an attempt at data manipulation.

Sources: [2] [5] [6]

Further reading:

Test example: a 1.5 percent change in result

The following example serves as an internal reference guide. It does not describe any completed product testing or analytical release limit. Example dataset Q17 has the documented prior result R1 with an area value of 100,000 arbitrary units. The laboratory intends to perform a method change and preliminarily sets a detection threshold above 1.0 percent.

The threshold is used here exclusively to facilitate targeted evaluation of comparative results. It is neither a drug product specification nor a general software tolerance. For each individual case, a controlled copy of the identical initial state is utilized. The table defines the expected outcome; actual observations are entered only upon execution.

Example: Scenario Q17 with separate actions and independent calculations
SubcaseSpecified reference valuesSoll and Decision
Only display the result R1 = 100,000; displayed reference value = 100,000 0% Difference; verify identity and original context additionally.
Rerun the workflow R1 = 100,000; assumed R2 = 101,500 (101,500 − 100,000) / 100,000 × 100 = 1.5 percent; investigate root cause.
Edge case of own regulations Accepted reference value = 101,000 1.0 percent; the rule "more than 1.0 percent" has not been triggered. Further acceptance criteria remain valid.
Wrong specimen despite same value Q18 instead of Q17; both values = 100,000 Assignment review not passed; numerical equality is insufficient.
Export results Q17/R1 with specified required supporting information Demonstrate data and method traceability, as well as required history, at the intended point of review.

Sources: [2] [5]

User, editing permissions, and audit scope are defined

Prepare a small activity matrix: measure, modify methods, evaluate, verify results, export, and administer. Describe the permitted actions for each role. Verify subsequently with representative accounts whether the actual permissions align with the defined scope. A role designation alone does not constitute effective restriction.

The LC/TQ Release Notes list in the historical section 12.0 audit functions for methods, worklists, data files, and studies. This specific reference assists in categorizing existing installations. It does not establish a complete event log for all MassHunter versions. For your system, it must be demonstrable which changes to specific objects can be documented and traced.

Plan the review such that relevant changes and their impact on the outcome are clearly visible. Frequency and scope should be determined based on the process and risk. If electronic signatures are used, additionally define their significance and data set reference. A login, an edit history, or a stored file does not automatically fulfill this requirement.

Sources: [3] [2] [5]

Restore the required data linkage

Determine which information enables traceability of a laboratory test result within the intended retention period. This may include measurement data, methods, relevant result statuses, processing information, and assignment to the corresponding task. Define where these information items are located and how an authorized reviewer can access them.

An appropriate recovery scenario begins with a defined laboratory incident, not merely with a successfully copied file. Following restoration, the required information must be readable, properly linked, and professionally usable. During evaluation, also assess the required software or service accessibility. The specific selection and depth should be justified based on risk assessment.

During a data or format migration, specify what must be fully available at the target site and what remains under control in the legacy system. Avoid blanket decisions regarding deletion or retention based solely on file age. Key considerations include the intended purpose and the applicable deadlines.

Sources: [5] [6] [2]

Assess the impact of updates

Classify a planned change to one of the following levels first: Device Registration, Evaluation Algorithm, Method Inventory, Data Format, Storage Path, or User Management. This results in different validation evidence. A change to the evaluation algorithm requires different comparison cases than a simple update to the authorized user group.

Evaluate release notes, compatibility, and existing supplier tests against your requirements. Appropriate evidence may be adopted. Supplement any process transitions that do not adequately cover the required areas. A documented justification is more helpful than automatic replication of all previous tests or an unexplained decision to omit them.

In addition to operational control, you shall manage unexpected results and technical incidents. Secure the necessary source information, evaluate relevant decisions, and define appropriate follow-up actions. A repeat run or re-analysis should be justified by a clear and traceable rationale, rather than merely serving to generate a desired outcome.

Sources: [2] [3] [5]

What Laboratory, Quality Assurance, and IT should prepare

For a small company, a shared inventory review eliminates unnecessary loops. The laboratory provides typical methods, data sets, and decision rules. IT defines components, accounts, and data pathways. QA and the responsible process owners establish the defined purpose, the required evidence, and the handling of open findings.

Charges consist of different device families, multiple analysis stages, unclear historical data linkages, and changes to central services. Available test data and the alignment of existing supplier documentation also influence the scope. Only after defining boundaries can a reliable planning approach be established; the product name does not provide a credible overall duration.

  • Have all components been captured with build, device, and valid compatibility basis?
  • Are Display, Reprocessing and manual editing separately described?
  • Are source data, method status, and expected results prior to testing defined?
  • Are critical assignment, authorization, and error scenarios included?
  • Is the required context for export and restoration verifiable?
  • Are professional assessment, operational responsibility, and changes governed?

Sources: [5] [2] [3]

Project experience and collaboration

Link measurement data, analysis and decision

With your laboratory, we follow a MassHunter analysis from measurement to the conclusion used. We connect the method version, input data, processing and report. A single selected method may justify a focused test package. For a new laboratory application, process owners and IT first establish what information supports the result and must remain available for reassessment. We then use a known dataset to check retrieval and traceability of changes.

Frequently asked questions about Agilent MassHunter

Can Agilent MassHunter be validated as a single system?

You may define a common validation scope, but it must include a detailed description of the actual components and transfer points. LC/TQ data acquisition, qualitative and quantitative analysis, and data retention have distinct responsibilities and versions. A cohesive validation evidence is advisable when boundaries remain clearly visible. A single version statement or a general manufacturer certificate does not replace this detailed description.

Sources: [1] [2] [3] [5]

Is Qualitative Analysis 14.0 compatible with every Acquisition version?

No, such a general compatibility is not documented. The release notes call specific dependencies. The separately viewed LC/TQ-12.3 documentation describes an older connection with Qualitative 13.0. Set the appropriate manufacturer stands for the planned combination. Unexplained combinations must be clarified before use, even if both programs can be installed.

Sources: [2] [3]

Does MassHunter require OpenLab CDS or ECM?

No. The documented operating modes differ between local workstations, networked workstations, and a specific OpenLab-CDS client/server context. The central services required depend on the component and its configuration. Document the actual data and login pathway. An installation with the family name MassHunter does not currently include central archiving or a specific signature function.

Sources: [2] [3]

Why might a previous result differ after an update?

First, determine whether the previous result was merely displayed or recalculated. Qualitative 14.0 outlines potential algorithmic effects arising from repeated processing. Compare identical input data and traceable method versions. A discrepancy requires professional evaluation; it is neither automatically indicative of a software defect nor automatically acceptable. The intended analytical decision determines the appropriate course of action.

Sources: [2] [5]

Can manual integration be included as a normal test case?

Yes, if it pertains to your actually authorized evaluation process. Describe the reason, authorized role, specific action, and required evidence. Operation and history must be documented for the evaluation component in use. This guide does not assume identical integration functionality across all MassHunter programs. Assess the original result and revised processing within the intended review context.

Sources: [2] [5] [6]

Does a .d-export prove the complete transfer of a .sirslt-data set?

No. The documented export capability indicates that this transfer pathway exists. It does not automatically ensure identical audit information, metadata, or signature links. Prior to implementation, define the required information scope and compare source to destination. If relevant information is controlled and remains at the source, access and assignment mechanisms must continue to function for the intended purpose.

Sources: [2] [5]

Is a PDF report sufficient for subsequent evaluation of the results?

A PDF results report alone does not equate to the complete measurement and analysis dataset. It summarizes selected information. Obtain the electronic data together with the required method version and relevant processing history. EU GMP Chapter 4 requires the definition of key raw data; this includes at least the data upon which quality decisions are based. At a representative result, verify that origin, method, and subsequent changes remain traceable.

Sources: [6] [5]

Are all available audit trails fully reviewed for each measurement run?

A blanket comprehensive approach is not a practical requirement. Determine, based on the process and risk, which changes are relevant to result evaluation and when they should be reviewed. For this, the actual existing information sources must be known. Historical LC/TQ data for methods and worklists does not substitute for an up-to-date event log of the installation or a justified review procedure.

Sources: [3] [5]

Must every test be re-performed after each MassHunter update?

Not automatically. Assess the affected components, functions, data, and existing evidence. Comparative evaluations may be significant following algorithm changes; alternative reviews should be prioritized following changes to access rights. Document the selection and outstanding risks. Appropriate supplier testing is applicable when its actual coverage aligns with the intended system and your requirements.

Sources: [2] [5]

Which documents assist in the initial cost estimation?

The following are required: component and device lists, versions, data pathways, roles, representative methods and data sets, as well as existing validation and supplier documentation. Planned changes and known incidents should be included. This enables visibility into open handovers and facilitates assignment of audit tasks. Without this foundation, fixed statements regarding project duration or test volume would remain assumptions, not a reliable plan.

Sources: [5] [2] [3]

Sources and review date

As of 6 October 2026. Product references: MassHunter Qualitative Analysis 14.0, May 2026, and LC/TQ Acquisition 12.3, May 2025, including explicitly historical subsections. These versions do not constitute a pre-approved combined configuration. Quantitative integration details are to be clarified on a version basis. Tables and numerical data are for internal reference purposes only; no installation has been tested for this purpose.

  1. Agilent MassHunter Acquisition for LC/MS
  2. MassHunter Qualitative Analysis 14.0 – Release Notes
  3. MassHunter Acquisition for LC/TQ 12.3 – Release Notes
  4. European Commission: EudraLex Volume 4
  5. EU GMP Annex 11: Computerised Systems
  6. EU GMP Chapter 4: Documentation
  7. FDA: Part 11 — Scope and Application

Sources checked: .

Daniel Herrmann, Daniel Herrmann Consulting

Daniel Herrmann Consulting

Provide validation evidence for your LC/MS process

We support you in segregating components and data pathways, evaluating existing documentation, and preparing credible testing protocols for your laboratory process.

  • Computer system validation →

    Bring your GxP systems into validated operation, from laboratory software, MES and eQMS to ERP. We define the scope, plan risk-based tests using GAMP 5 and hand the evidence to QA for release.

Send a project enquiry Customer testimonials →