CanRisk vs Evagene: risk calculator versus integrated pedigree platform

Short version. CanRisk is the authoritative University of Cambridge web front end to BOADICEA (breast and ovarian cancer) and CanRisk-Prostate. It is backed by Cambridge, Cancer Research UK, the Wellcome Trust, and Genome Canada, and endorsed by NICE, NCCN, ESMO, Cancer Care Ontario, eviQ, and the UK Cancer Genetics Group. It is the tool that clinical services reach for when they need a defensible, current, endorsed BOADICEA risk number — and that is a capability Evagene does not and should not try to replace. Evagene is a clinical pedigree management system whose cancer-risk layer is built on BRCAPRO, MMRpro, and PancPRO from the BayesMendel suite, alongside Mendelian calculators, AI-assisted interpretation, and a platform surface for EHR and AI-agent integration. Positioned honestly, CanRisk and Evagene are complementary: CanRisk runs BOADICEA; Evagene manages the pedigree workflow, runs other models, drafts the interpretation, and hands the data on to the rest of the clinical system. Most cancer genetics services already run CanRisk alongside a separate pedigree platform; Evagene is designed to sit in that second slot.

This article is a careful, respectful comparison written for service leads evaluating tooling in 2026. All claims about CanRisk are drawn from the canrisk.org public website, the BOADICEA/CanRisk peer-reviewed literature, and the official endorsements listed on the CanRisk site as of April 2026.

How the two products position themselves

CanRisk describes itself as an online tool that enables healthcare professionals to calculate an individual's future risks of developing breast, ovarian, or prostate cancer. The emphasis is explicitly on risk calculation. The tooling sits behind a professional registration wall — authorised healthcare professionals only — which is appropriate given the clinical sensitivity of the numbers it produces. Its model stack is serious and current: BOADICEA (Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm) for breast and ovarian risk, CanRisk-Prostate for prostate risk, with polygenic risk scoring, mammographic density integration, family history, and results from genetic testing. The tool is available in seven languages. It enjoys NICE, NCCN, ESMO, Cancer Care Ontario, eviQ, and UK Cancer Genetics Group endorsement — the most comprehensive endorsement set of any tool in this space, and one that reflects the weight of the Cambridge academic work behind BOADICEA over two decades.

What CanRisk is not, by its own positioning, is a general-purpose pedigree management system. It is a risk calculator with a pedigree input. The pedigree there is a means to the end of producing a risk number; it is not where you manage ongoing clinical notes, batch-screen for non-cancer conditions, cascade test across a family, or draft a clinical letter. Those things are done elsewhere. The clinical-service norm is to run CanRisk at the risk-assessment moment and run a separate pedigree tool for the rest of the workflow.

Evagene is that separate pedigree tool. It positions itself as clinical-grade pedigree management for precision medicine: a gesture-drawing canvas for consultation-speed construction, NSGC/ISCN-standard notation, a curated 200+ disease catalogue annotated with ICD-10 and OMIM codes, and validated risk calculators for conditions outside BOADICEA's scope (Lynch syndrome via MMRpro, pancreatic cancer via PancPRO, and Mendelian AD/AR/XR for monogenic non-cancer conditions). On top of that sit the modern capabilities that make Evagene useful as a platform: AI-powered clinical interpretation using bring-your-own-key (BYOK) LLMs (Anthropic Claude, OpenAI GPT, Fernet-encrypted at rest), Analysis Templates, a REST API with scoped evg_ keys, webhooks (HMAC-SHA256), an MCP server exposing 11 pedigree tools to AI agents like Claude Desktop and Claude Code, and an embeddable pedigree viewer.

The headline difference is of shape. CanRisk is a deep, authoritative, narrow-scope calculator for specific cancers. Evagene is a broad clinical pedigree platform that happens to include risk models for conditions BOADICEA does not cover.

Feature-by-feature comparison

The matrix below is arranged by capability. A tick means the feature is publicly documented. Because CanRisk and Evagene overlap only partially, many rows will have the tick on one side and a dash on the other; that is a feature of the comparison, not a flaw. The point is to show that the two products complement each other cleanly.

Matrix compiled from canrisk.org and Evagene public product pages as of April 2026. A dash does not imply absence; it indicates the capability is not publicly advertised.

BOADICEA and BRCAPRO: different models, both valid

A frank word on the risk models. BOADICEA — the engine behind CanRisk — and BRCAPRO — integrated in Evagene — are both peer-reviewed models for estimating breast and ovarian cancer risk and BRCA1/BRCA2 mutation probability. They differ in design. BRCAPRO is a Bayesian family-history-driven model from the BayesMendel suite; it uses family history of breast and ovarian cancer with published penetrance estimates for BRCA1 and BRCA2. It is validated, widely used in US practice, and computationally tractable. BOADICEA is a more comprehensive model that integrates family history, mutations in multiple high- and moderate-risk genes, polygenic scoring, mammographic density, and hormonal/lifestyle factors. BOADICEA is generally regarded as the more complete modern model; it is what NICE, NCCN, ESMO, and others point clinicians toward.

Evagene does not attempt to re-implement BOADICEA. Doing so would be an enormous piece of work, and the clinical-equivalence validation work that Cambridge has done over two decades cannot be casually reproduced. The honest statement is: if your service needs BOADICEA numbers, you use CanRisk. Evagene's BRCAPRO output is complementary — useful for cross-checking, useful in services where BOADICEA is not the standard, useful for services in jurisdictions where BRCAPRO is the referenced model — but it is not a substitute for the gold-standard endorsed tool.

Where Evagene adds value on the cancer side is in the conditions BOADICEA does not cover. Lynch syndrome risk, handled in Evagene by MMRpro, sits outside BOADICEA's scope. Pancreatic cancer risk, handled in Evagene by PancPRO, sits outside BOADICEA's scope. And non-cancer monogenic conditions — covered in Evagene by Mendelian AD/AR/XR calculators — sit outside CanRisk entirely, because CanRisk is a cancer tool.

Pedigree management: where CanRisk stops

CanRisk is authoritative on the risk number but deliberately narrow on the workflow around it. It does not aim to be a pedigree management system; it does not store a family's clinical history over time, draft correspondence, manage cascade testing across relatives, batch-screen a proband against a wider disease catalogue, or expose the pedigree to other systems via an API. Clinical services that use CanRisk typically pair it with a pedigree tool — historically Progeny, Phenotips, TrakGene, FamGenix, a spreadsheet, or a paper chart — and move the family structure between tools via the CanRisk file format or manual re-entry.

Evagene is designed to be the pedigree tool in that pairing. Its gesture drawing mode supports live construction during consultation. Its 200+ disease catalogue covers cancer and non-cancer conditions, so a single pedigree covers the family's full history rather than just the BOADICEA-relevant fields. Its AI interpretation layer drafts the narrative paragraphs a clinician will later adapt and sign, leaving the BOADICEA risk number (pulled from CanRisk) to be pasted alongside. Its REST API, webhooks, and MCP server mean the pedigree is visible to other systems — a patient portal, an EHR, an AI agent — without bespoke integration work.

Using CanRisk and Evagene together

The intended workflow is simple. The counsellor builds the pedigree in Evagene during consultation — gesture drawing, live annotation, disease coding against ICD-10 and OMIM. At the risk-assessment moment, the relevant pedigree structure and tumour pathology are transferred to CanRisk (CanRisk file format from Evagene export, or manual re-entry for small families). CanRisk produces the authoritative BOADICEA / CanRisk-Prostate risk number. That number goes into the Evagene case notes alongside the BRCAPRO, MMRpro, and PancPRO results from Evagene's own engine. The AI interpretation layer drafts the report, cross-referencing the CanRisk BOADICEA figure as the authoritative breast-ovarian risk. The clinician reviews, edits, and signs the final report in Evagene; that report is exported to the EHR via the REST API or a webhook.

For services that are already CanRisk-fluent, this is a small change. You add Evagene alongside your existing CanRisk workflow; CanRisk's role is unchanged. What Evagene replaces is whatever you are currently using for the pedigree itself — Progeny, Phenotips, TrakGene, FamGenix, or a spreadsheet. See our comparisons with each of those tools for specifics: Progeny, Phenotips, TrakGene, and FamGenix.

When to choose CanRisk (and only CanRisk)

• Your service's central question is the BOADICEA risk number, and the pedigree is a means to that end.
• You are working in breast, ovarian, or prostate cancer risk only.
• You need polygenic risk score integration, mammographic density, or current BOADICEA methodology.
• Your institutional protocol references BOADICEA / CanRisk-Prostate specifically (per NICE or other guideline).
• You are comfortable managing the wider pedigree workflow (notes, cascade, correspondence) in your existing tooling.

When to choose Evagene (alongside CanRisk)

• You need a pedigree platform that covers cancer and non-cancer conditions, not just BOADICEA-relevant fields.
• You want BRCAPRO / MMRpro / PancPRO running directly on the pedigree as a complement to CanRisk.
• AI-assisted report drafting using your own LLM keys is part of how your service wants to work.
• You need programmatic pedigree access (REST, webhooks, MCP, embeddable viewer) for EHR and patient-portal integration.
• You want GEDCOM, JSON, 23andMe, and image-OCR import to move data in and out of the pedigree freely.
• You need batch risk screening across a 200+ disease catalogue for non-BRCA workflows.

Migration path: bringing pedigrees between the tools

CanRisk uses its own CanRisk file format (a structured CSV) for pedigree and tumour-pathology input. Evagene exports GEDCOM 5.5.1 and JSON. A conversion between JSON or GEDCOM and the CanRisk file format is short to write and can be automated; for occasional use, manual re-entry of the relevant subset (individuals, sex, year of birth, affected status, cancer type and age of diagnosis) is practical for small pedigrees. Polygenic score, mammographic density, and explicit genotype fields are typically entered directly in CanRisk because they are BOADICEA-specific. The GEDCOM pedigree software guide explains the format-conversion landscape in more detail.

Going the other way — from CanRisk to Evagene — is also practical. The CanRisk file format can be parsed to extract the pedigree structure, which Evagene's JSON import accepts. Tumour-pathology fields map to Evagene's disease annotations with ICD-10 coding; cancer diagnoses with age at diagnosis translate cleanly.

Frequently asked questions

Related comparisons and reading

• Phenotips vs Evagene
• Progeny vs Evagene
• FamGenix vs Evagene
• TrakGene vs Evagene
• Hereditary cancer risk assessment
• Clinical genetics pedigree tools