Saved Results

No results saved yet.

Enter a patient name and hit Save on a result.

Clinical calculator summary

DLBCL Prognosis Panel

Strict pretreatment interpreter of five separate source-locked prognostic indices.

Evidence-based context for fast calculator use

Purpose:
Present Classic IPI, NCCN-IPI, aaIPI, CNS-IPI, and R-IPI without combining their endpoints.
Population:
Adults with newly diagnosed de novo DLBCL at pretreatment baseline; component treatment and age restrictions apply.
Factors:
Age, Ann Arbor stage, ECOG, Raw LDH/ULN ratio, Extranodal count, Source-specific extranodal sites
Reference:
Original IPI/aaIPI (1993), R-IPI (2007), NCCN-IPI (2014), and CNS-IPI (2016).
HomeDLBCL Prognosis Panel
Pin your most used calculators here by clicking the star in the dropdown.

DLBCL Prognosis Panel

U/L

Enter a value greater than 0 from the same laboratory as the ULN.

U/L

Enter a value greater than 0. The raw LDH/ULN ratio is used without rounding.

Clinical Context & Background

This panel reports five separate prognostic models for adults with newly diagnosed de novo diffuse large B-cell lymphoma (DLBCL) at pretreatment baseline. It does not average, rank, reconcile, or select a final model.
Classic IPI is the original 1993 aggressive non-Hodgkin lymphoma model. Its 5-year overall-survival percentages are historical original-cohort observations. NCCN-IPI was developed in adults with de novo DLBCL in the rituximab era. Age-adjusted IPI (aaIPI) is reported only for age 60 years or younger. R-IPI is reported only for an R-CHOP context. CNS-IPI is reported only for R-CHOP or R-CHOP-like treatment, age 80 years or younger, and no known CNS disease at baseline.
Each endpoint, cohort observation, applicability decision, and calculation trace remains separate. Component not-applicable status is never scored as zero. Cohort percentages are observations from study populations, not calibrated patient-specific predictions and not treatment rules.
Primary Source Evidences: https://doi.org/10.1056/NEJM199309303291402; https://doi.org/10.1182/blood-2006-08-038257; https://doi.org/10.1182/blood-2013-09-524108; https://pmc.ncbi.nlm.nih.gov/articles/PMC5527396/; https://doi.org/10.1200/JCO.2015.65.6520.
Formula Logic
Strict shared baseline contract with semantic adapters to verified Classic IPI and NCCN-IPI engines plus source-locked aaIPI, CNS-IPI, and R-IPI components. No combined score, ranking, or model averaging.

Reference Data

Panel componentSource-locked outputSynthesis boundary
Component resultsEach verified engine retains its own score, status, trace, cohort, and endpointNo averaging, ranking, or merged clinical category
Unavailable componentReported as unavailable or not applicableNever converted to zero or a favorable result

Clinical Workflow

Use, Interpret, And Continue The Patient Pathway

Expand for workflow guidance, limitations, examples, and related next steps.

When To Use

  • After de novo DLBCL diagnosis and before first-line treatment.
  • When the inputs required by the model components you want to review are explicitly known.

How To Interpret

  • Read each applicable model separately with its endpoint and cohort.
  • Treat component N/A as outside that model scope, never as zero.

What To Do Next

  • Integrate results with pathology, molecular findings, patient factors, and clinical judgment outside this calculator.

Limitations

  • No model is averaged, ranked, or selected as a final answer.
  • Historical cohort percentages are not calibrated patient-specific probabilities.
  • The panel does not recommend treatment or CNS prophylaxis.

Validated Population

Implementation scope: adults with newly diagnosed de novo DLBCL at pretreatment baseline.

Calculation verification history

Verified

  1. : Production browser revalidation

    • Three source-backed production cases passed through visible controls across the all-zero baseline, the NCCN-IPI age-41 transition, and a high-risk profile with age-adjusted IPI correctly not applicable above age 60.
    • Observed component results matched exactly: 0/0/0/0/0 at baseline, NCCN-IPI 1 with all siblings unchanged at age 41, and Classic IPI 5, NCCN-IPI 8, CNS-IPI 6, R-IPI 5 with aaIPI not applicable in the maximum case.
    • The published panel presents each model as an Independent result with no agreement/discordance synthesis, combined score, stale controls, literal undefined/null values, or visible error boundary.
    Download latest production check (Excel)

    Case-level production evidence is retained in validation/browser-revalidation/panel-lymphoma-production-closure-2026-08-03.json.

  2. : Local independent-component summary remediation

    • Three visible-control cases covered the age-40 baseline, the exact age-41 NCCN-IPI transition, and the maximum-applicable high-risk profile.
    • The concise receipt now labels the second column `Independent result` and contains no agreement/discordance instruction or cross-model synthesis.
    • All five component outputs retained their source-specific scores and applicability; 1,536/1,536 assertions passed and the browser console had no errors.

    Case evidence is retained in validation/browser-revalidation/panel-lymphoma-local-remediation-2026-08-03.json.

  3. : Production browser revalidation — UI finding retained

    • Three source-backed cases matched all independent Classic IPI, NCCN-IPI, aaIPI, CNS-IPI, and R-IPI results at baseline, the age-41 NCCN transition, and a maximum applicable profile.
    • Production still labels the panel summary `RESULT / AGREEMENT` and instructs users to `review agreement or discordance clinically`, outside the neutral parallel-results contract.
    • Overall production status remains failed-ui-safety until that residual synthesis wording is removed and revalidated; no browser-console error was observed.

    Case-level evidence and reproduction are retained in validation/browser-revalidation/panel-lymphoma-2026-08-03.json.

  4. : Local browser remediation verified; production retest pending

    • Ten source-backed cases passed through visible controls on the corrected local route across 11 submissions.
    • Classic IPI, NCCN-IPI, aaIPI, CNS-IPI, and R-IPI remained separate across thresholds, treatment contexts, known CNS disease, outside-scope, and reset/repeat paths.
    • No unified score, combined assessment, agreement/discordance synthesis, ranking, or preferred model was displayed. The original production findings remain open until deployment and public revalidation.

    Case-level local remediation evidence is retained in validation/browser-production/panel-lymphoma.local-remediation.json.

  5. : Production browser validation

    • Ten source-backed cases were exercised on the public production route through visible browser controls; 1/10 case records passed their expected-versus-observed assertions across 11 visible submissions.
    • Overall browser status: failed-ui-safety-and-console. critical: All nine rendered component-result cases were headed `Unified Panel Results`, described as a `Combined assessment from multiple tools`, and instructed users to compare `agreement and discordance alongside the unified panel result`. | medium: Production emitted minified React error #418 during route loads/reloads used by this run.
    • Case-level inputs, expected and observed results, reset/repeat behavior, failure reproductions, and console evidence are retained in the production-browser evidence record.

    Case-level production evidence is retained in validation/browser-production/panel-lymphoma.json.

  6. : Exact input guidance for unavailable components

    • Skipped, incomplete, invalid, and not-applicable component explanations now expand model language, use "Please review", and identify exact visible entries in parentheses.
    • Treatment-context explanations spell out the regimen while pointing back to (DLBCL baseline context).
    • No score, threshold, cohort outcome, applicability rule, or component-isolation behavior changed.

    All 1,536 existing panel assertions pass; current report remains applicable.

  7. : Input workflow simplified and revalidated

    • Reduced the public form to result-, branch-, unit-, and applicability-changing inputs; removed or consolidated non-scoring, administrative, and duplicate questions.
    • Preserved the verified calculation or report-interpretation logic, thresholds, model version, component logic, and supported result pathways behind strict deterministic adapters.
    • Passed the calculator-specific regression audit, adapter-parity and malformed-input checks, the 144-calculator input-quality audit, and the 43-calculator entry-burden audit.

  8. : Input-quality and partial-result verification

    • Marked diagnosis and pretreatment timing as the shared scope inputs, documented score-specific optional fields, and required strictly positive LDH and LDH upper-limit values.
    • Added scope-first handling plus explicit skipped, invalid, and not-applicable component states so blank optional groups do not block the panel and unaffected sibling scores remain available.
    • Added partial-entry, contradictory-input, component-isolation, and sibling-preservation fixtures while retaining the five source-locked scoring formulas and complete-case parity checks.

  9. : Full validation test

    • 1,536/1,536 source, strict-contract, component-parity, boundary, translation, applicability, orchestration, provenance, history, report, and evidence assertions passed.
    • This is implementation verification, not independent clinical validation; all five model endpoints and cohort observations remain separate.

  10. : Correction

    • Replaced duplicated inline IPI, NCCN-IPI, and R-IPI logic with verified child-engine adapters and a pure source-locked panel engine.
    • Removed defaults, coercion, caps, rounded-ratio decisions, aaIPI use above age 60, baseline-CNS CNS-IPI misuse, mixed endpoint highlighting, treatment directives, and model-preference claims.

Verification confirms the calculator implementation against the cited model; it is not independent clinical validation, regulatory approval, or medical advice.

Found a possible issue? Report it to improve this calculator

Share This Calculator

Share:

Related Tools

Myeloma R2-ISS and IMWG Frailty PanelFebrile Neutropenia TriagePalliative Survival Prediction

Frequently Asked Questions

When should DLBCL Prognosis Panel be used?

Use it for patients whose clinical question requires the component models displayed in the combined panel when all required inputs and the intended clinical setting are confirmed.

Can DLBCL Prognosis Panel determine treatment by itself?

No. Interpret the result with the cited evidence, complete clinical assessment, current guidelines, and patient-specific goals.

Evidence-based oncology decision support. Verify with clinical guidelines.