Prostate cancer architectural grading & ISUP risk stratification simulator
Every Gleason score begins with tissue in hand. A systematic needle biopsy samples the prostate at defined anatomic sites, and standardized histologic processing converts that tissue into a slide the pathologist can actually read. Sampling error and processing artifact are the two biggest threats to an accurate grade before a single pattern is even assessed.
A systematic prostate biopsy places 10–12 needle cores through defined regions of the gland (apex, mid, base, bilaterally, plus lateral/peripheral zone sampling), often guided by transrectal ultrasound or multiparametric MRI fusion targeting. Each thin core of tissue is placed in a labeled cassette, fixed in 10% neutral buffered formalin for structural preservation, then dehydrated and embedded in paraffin wax (formalin-fixed paraffin-embedded, "FFPE").
A microtome cuts the paraffin block into 4 µm ribbons, which are mounted on glass slides and stained with hematoxylin (stains nuclei blue-purple) and eosin (stains cytoplasm and stroma pink) — the workhorse stain of surgical pathology for over a century.
Unlike most cancer grading systems that focus heavily on individual cell/nuclear features (nucleolar prominence, mitotic count, pleomorphism), the Gleason system is almost entirely an architectural grade — it is read at low-to-medium magnification (4x–10x objective) by evaluating how glands are organized and arranged relative to one another and to the surrounding stroma, not by scrutinizing single-cell atypia.
This was a deliberate design choice by Gleason: architectural pattern proved more reproducible between pathologists and correlated more strongly with clinical outcome than nuclear grading schemes tested at the time. It remains one of the most validated prognostic systems in all of oncology.
The original Gleason study (1966–1974 Veterans Administration Cooperative Urological Research Group) evaluated over 4,000 patients — one of the largest grading validation studies ever performed, which is part of why the system has remained the backbone of prostate cancer pathology for nearly 60 years.
Because each core samples only a needle-thin sliver of a walnut-sized gland, a biopsy can under-sample the true highest-grade focus of a tumor — a phenomenon called grade migration or sampling error. Studies comparing biopsy Gleason score to the score found at radical prostatectomy show upgrading (final grade higher than biopsy) in roughly 30–45% of Grade Group 1 biopsies, which is precisely why active surveillance protocols schedule confirmatory re-biopsy within the first 12 months.
MRI-targeted biopsy of visible lesions (PI-RADS ≥3) alongside systematic sampling has meaningfully reduced — though not eliminated — this under-grading problem.
The heart of Gleason grading is a five-tier catalogue of glandular architecture, from perfectly formed individual round glands (pattern 1) to sheets of cells with no gland formation whatsoever (pattern 5). Move the sliders to watch each pattern render live in the tissue view — notice how gland shape, fusion, and lumen formation degrade as the pattern number rises.
Pattern 1: small, uniform, closely packed round-to-oval glands with a single layer of epithelium, occupying a well-circumscribed nodule — architecturally almost indistinguishable from benign tissue at a glance. Pattern 2: similar round glands but slightly more variable in size and more loosely arranged, with more intervening stroma and slightly infiltrative margins.
Both patterns are now rarely, if ever, diagnosed on needle biopsy. The 2005 and 2014 ISUP consensus conferences concluded that what earlier pathologists called pattern 1–2 cancer was often actually benign mimicker tissue (e.g., adenosis) or represented such an indolent, non-infiltrative process that assigning it as cancer at all was reconsidered. Today these patterns appear almost exclusively in historical literature or in glands from transurethral resection specimens.
Pattern 3 is now the lowest pattern realistically assigned on contemporary biopsy: individual, discrete glands of variable size, haphazardly infiltrating between benign glands and through the stroma, each still forming a recognizable single lumen lined by a single epithelial layer. The key discriminator from pattern 1–2 is infiltrative growth — the glands invade individually rather than growing as a cohesive nodule.
Pattern 3 comprises the majority of "low-risk" prostate cancer diagnosed today (Gleason 3+3=6, Grade Group 1).
Pattern 4 marks a qualitative jump: individual gland lumina are lost through fusion. Glands merge into cribriform structures (a sheet of cells punched through with multiple round lumina, resembling a slice of Swiss cheese, with no intervening stroma), glomeruloid formations (glands that fold in on themselves resembling a renal glomerulus), or poorly-formed glands (ragged, incomplete gland outlines with only occasional lumina).
Pattern 5 represents essentially complete loss of glandular differentiation: solid sheets or nests of tumor cells, single infiltrating cells, or cords, with no gland lumina at all. Comedonecrosis — central necrotic debris within a solid nest or cribriform gland — is a specific, high-yield feature that automatically qualifies an area as pattern 5 regardless of the surrounding architecture.
Because grading is a continuum, most tumors show more than one pattern side by side in the same core — which is exactly why the system requires reporting a primary AND a secondary pattern rather than a single number.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Pattern 1 | Small, uniform round glands, single layer epithelium | Tightly packed, circumscribed nodule, minimal stroma | Historical only — not diagnosed today |
| Pattern 2 | Round glands, more variable size | Slightly more spaced, mildly infiltrative margin | Rare — mostly historical/TURP specimens |
| Pattern 3 | Discrete infiltrative single glands, variable size | Individual glands invade between benign glands | Modern floor of biopsy-diagnosed cancer |
| Pattern 4 | Fused / cribriform / glomeruloid / poorly formed glands | Lumina fuse or are lost; sheets with punched-out holes | Defines intermediate-risk disease |
| Pattern 5 | Solid sheets, cords, single cells, ± comedonecrosis | No gland lumina form at all | Defines the highest-risk architecture |
A Gleason score is not a single pattern — it is a sum of the two most prevalent architectural patterns observed anywhere in the tumor. The primary pattern is whichever pattern occupies the greatest area (>50% of the tumor); the secondary is the next most prevalent. Their addition, written as "primary+secondary=sum" (e.g. 3+4=7), preserves information a single number would erase.
The pathologist estimates the percentage of tumor occupied by each pattern present across all positive cores (or, for a single core / focus, within that specimen). The pattern making up the largest area becomes the primary number; the next largest becomes the secondary number. These are simply added: 3 (primary) + 4 (secondary) = Gleason Score 7.
Critically, 3+4=7 and 4+3=7 are NOT the same disease. Both sum to 7, but a tumor that is mostly well-formed pattern 3 with a minor pattern 4 component behaves very differently from one that is mostly poorly-formed pattern 4 with only a minor pattern 3 component. This asymmetry is exactly why the Grade Group system (Stage 4) was created — sum alone hides which pattern dominates.
If a tumor shows only one architectural pattern throughout, that pattern is simply doubled (a purely pattern-3 tumor is scored 3+3=6, not just "3"). On radical prostatectomy specimens (as opposed to biopsy), when a third, higher-grade pattern is present but occupies less than 5% of tumor volume, it is recorded separately as a "tertiary pattern" (e.g., "Gleason 3+4=7 with tertiary pattern 5") because even a small high-grade component carries independent prognostic weight.
On needle biopsy, if a third pattern is present, the standard convention differs slightly: the highest-grade pattern is generally incorporated as the secondary pattern regardless of its percentage, since biopsy tissue volume is limited and under-representing the worst pattern present is considered a greater clinical risk than over-grading.
A tumor described as "4+3=7" is graded Grade Group 3, while "3+4=7" — the identical numeric sum — is Grade Group 2, with roughly a 25-percentage-point difference in 10-year biochemical recurrence-free survival between them. The sum alone would never reveal this.
When multiple biopsy cores are positive, each core is typically graded individually, and an overall (or "global") Gleason score for the case is also assigned reflecting the highest-grade / most clinically significant core, or a composite of the overall pattern distribution across all positive cores, per current College of American Pathologists and ISUP reporting protocols. Reporting both the per-core grades and the overall case grade gives the treating physician a fuller picture of tumor volume and grade heterogeneity across the gland, which matters for treatment planning (e.g., whether focal therapy is anatomically feasible).
In 2014, the International Society of Urological Pathology (ISUP) consensus conference — later adopted by the WHO in 2016 — collapsed the confusing landscape of Gleason sums into five Grade Groups with clearly separated, statistically validated prognoses. The system solved two problems at once: it separated 3+4 from 4+3, and it stopped patients from hearing "Gleason 6 out of 10" and assuming their disease was roughly in the middle of a severity scale.
Two problems motivated the change. First, prognosis-scrambling: Gleason 7 lumps together 3+4 (relatively favorable) and 4+3 (meaningfully worse) under one number, and the full clinical range historically ran from 2 to 10, even though scores below 6 were essentially never assigned in modern practice — leaving patients to (wrongly) interpret a "6" as roughly average severity on a 2–10 scale, when it is in fact the lowest-risk category diagnosed today.
Second, granularity beyond what mattered clinically: Gleason 8 (4+4), Gleason 9 (4+5 or 5+4), and Gleason 10 (5+5) all carry meaningfully different but overlapping prognoses that a raw sum does not clearly separate from one another or from Gleason 7 disease.
Grade Group 1: Gleason score ≤6 (essentially always 3+3) — tumors composed entirely of well-formed glands. Grade Group 2: Gleason score 3+4=7 — predominantly well-formed glands with a minor component of fused/cribriform/poorly-formed glands (traditionally <5–10% pattern 4, though any amount as secondary still qualifies). Grade Group 3: Gleason score 4+3=7 — predominantly fused/cribriform/poorly-formed glands with a minor well-formed component. Grade Group 4: Gleason score 8 — this includes 4+4=8, as well as 3+5=8 and 5+3=8. Grade Group 5: Gleason score 9–10 — 4+5, 5+4, and 5+5, the most architecturally disorganized tumors.
The ISUP Grade Group system was rapidly and near-universally adopted because it required no new tissue analysis — it simply relabels information pathologists were already reporting, while making prognostic separation immediately legible to both physicians and patients.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Grade Group 1 | Gleason ≤6 (3+3) | All glands well-formed | 10yr bRFS ≈ 96% |
| Grade Group 2 | Gleason 7 (3+4) | Mostly well-formed, minor fused/cribriform | 10yr bRFS ≈ 88% |
| Grade Group 3 | Gleason 7 (4+3) | Mostly fused/cribriform, minor well-formed | 10yr bRFS ≈ 63% |
| Grade Group 4 | Gleason 8 (4+4, 3+5, 5+3) | Predominantly poorly-formed/no-gland architecture | 10yr bRFS ≈ 48% |
| Grade Group 5 | Gleason 9–10 (4+5, 5+4, 5+5) | Near-total loss of glandular differentiation | 10yr bRFS ≈ 26% |
Grade Group, combined with PSA level and clinical/imaging stage, forms the backbone of NCCN and EAU risk stratification and drives the single most consequential decision in prostate cancer management: whether to watch, or to treat. The gap in outcome between Grade Group 1 and Grade Group 5 is one of the widest prognostic spreads of any solid-organ cancer grading system.
Grade Group 1 (Gleason ≤6) is now widely regarded as the preferred candidate for active surveillance rather than immediate treatment: regular PSA testing, periodic MRI, and repeat biopsy substitute for surgery or radiation unless the disease shows signs of progression, sparing most patients the sexual, urinary, and bowel side effects of definitive therapy for a cancer that may never become clinically significant in their lifetime.
Grade Group 2 (3+4=7) occupies a genuine gray zone: many contemporary protocols now extend active surveillance eligibility to select Grade Group 2 patients with low tumor volume (e.g., <10–20% of any core, limited number of positive cores) and favorable PSA density, while others recommend proceeding to definitive treatment — reflecting real, ongoing clinical debate rather than a settled rule.
Grade Group 3 (4+3=7) is generally treated with definitive therapy (radical prostatectomy or external beam/brachytherapy radiotherapy) in patients with reasonable life expectancy, given the meaningfully lower biochemical recurrence-free survival compared with Grade Group 2.
Grade Group 4–5 disease is classified as high-risk to very-high-risk under NCCN criteria and is typically managed with definitive local therapy combined with androgen deprivation therapy (ADT), often alongside radiotherapy, reflecting the substantially higher risk of occult micrometastatic spread at diagnosis. Multimodal treatment intensification (e.g., adding abiraterone or a second-generation antiandrogen to ADT plus radiotherapy) is increasingly used for Grade Group 5 disease based on modern randomized trial data.
Grade Group is only one axis of risk stratification — NCCN and EAU risk groups combine it with PSA level (e.g., >20 ng/mL raises risk regardless of grade) and clinical/imaging T-stage, so two patients with identical Grade Group 2 disease can land in different risk categories and receive different treatment recommendations.
Because grade is assessed from limited biopsy tissue, active surveillance protocols build in scheduled re-biopsy (commonly at ~12 months, then periodically thereafter) specifically to catch grade migration — cases where the true tumor grade was higher than the initial biopsy suggested. Roughly 25–30% of men on active surveillance are reclassified to a higher Grade Group within the first several years, most due to sampling correction rather than true biological progression, which is why grade is treated as an evolving assessment rather than a one-time verdict.