Invasive breast cancer

The article deals with invasive breast cancer and the evaluation of hormone receptor & HER2 status. Non-invasive breast cancer is dealt with in non-invasive breast cancer.

Introduction

Types of invasive breast cancer

Types:Ref.:

• Ductal - also known as no specific type (NST) - 79%.
• Lobular 10%.
• Cribriform (tubular) 6%.
• Mucinous (colloid) 2%.
• Medullary 2%.
• Papillary 1%.
• Metaplastic <1%.

Others:

• Angiosarcoma - post-radiation ~ 10 years.

Familial breast cancer

BRCA1 vs. BRCA2:

• BRCA1:
  • Younger.
  • Ovarian cancer.
  • Worse types of breast cancer (e.g. triple negative breast cancer: PR-, ER-, HER2/neu-).
• BRCA2:
  • Older.
  • Like sporatic.
  • Male breast cancer.
• BOTH are associated with increased risk of (memory device CPP):
  • Colon cancer.
  • Prostate.
  • Pancreas.

Breast IHC

Subtyping breast cancer

• DCIS vs LCIS:
  • E-cadherin (+ve DCIS, -ve LCIS).
  • antibody 34betaE12 (+ve perinuclear LCIS, -ve DCIS).
  • CAM5.2 (peripheral stain = DCIS, perinuclear stain = LCIS).
    • CAM5.2 is against CK8.
  • Beta-catenin (-LCIS, +DCIS).

• D2-40:
  • Monoclonal antibody to podoplanin.
  • Useful to assess lymphovascular invasion.

• ADH and DCIS:
  • E-cadherin.
    • Present in most epithelial cells.
    • Lost in LCIS & invasive lobular carcinoma.
  • SMMHC (smooth muscle cell myosin heavy chain).
    • Marks myoepithelial cells.

Treatment-related markers - overview

• Immunostaining of any sentinel lymph nodes - to look for isolated tumour cells and small lymph node mets.
  • Sunnybrook uses CAM5.2.
• ER (estrogen receptor).
  • Positive in most breast cancers; +ve in ~75-80%.
• PR (progesterone receptor).
  • Positive in most breast cancers; +ve in ~65-70%.
• HER2/neu.
  • Usually negative; -ve in 70-80%.
  • Positivity association with a worse prognosis.

ER & PR scoring

• Give a percentage, i.e. 0-100%.
  • Important cut points: 1% and 10%.
    • 0% = negative - not treated.
    • <10% = low positivity - treated.

Notes:

• Normal breast epithelial cells have a patchy staining for ER and PR.
• Evaluated on the invasive component.

HER2 scoring

Score	Staining intensity	Cells stained (%)	Membrane staining	Management	Percentage of cases
0	nil	<10%	incomplete	No HER2 blocker	~60%
1+	minimum	>10%	incomplete	No HER2 blocker	~10%
2+	weak	>10%	complete	Needs SISH or FISH	~10%
3+	strong	>10%	complete	HER2 blocker	~20%

Notes:

• Normal breast epithelial cells do not stain with HER2.
• Evaluated on the invasive component.
• SISH = silver in situ hybridization.
• FISH = fluorescence in situ hybridization.

Clinical

• ER & PR status determine whether a patient will get tamoxifen or other estrogen receptor modulators, such as raloxifene (Evista).
• HER2 status determines whether patient will get traztuzumab (Herceptin) or other HER2/neu modulators.

Characteristics of the subtypes

Ductal carcinoma

• AKA "NST" = No Specific Type.

Microscopic

Features:

• Cohesive cells - forming ducts or in sheets.
• Nuclear pleomorphism.

Clinical

• Typically: ER+, PR+, HER2-.

Lobular carcinoma

General

• May be associated with a CDH-1 mutation - seen in diffuse type gastric cancer.

Microscopic

Features:

• "Single file" - cell line-up in a row.
  • Cell should not be cohesive -- lymphoma should briefly come to mind.
    • primary lymphoma of the breast exists... but it is extremely rare.
• NO gland formation.
  • If it forms glands... it is more likely NST.
• May have signet ring morphology.
• NO desmoplastic reaction, i.e. the stroma surrounding the tumour cells should look benign and undisturbed.

Note:

• commonly have low grade nuclear features

Subclassification:

• Classic lobular carcinoma.
  • Low nuclear grade - NO significant variation of nucleus size.
• Pleomorphic lobular carcinoma.
  • Significant nuclear atypia.

Note: Some pathologist grade lobular carcinoma like other types and avoid the term "pleomorphic lobular carcinoma."

Medullary carcinoma

General

• Some pathologists don't believe this exists.

Epidemiology:

• Thought to have a better prognosis that no special type (NST).
• Association with BRCA1 mutations.

Microscopic

Features:

1. Lesion has well-circumscribed border.
2. Syncytial growth pattern = clumps of cells with poorly defined cell borders.
3. Lymphocytic infiltrate.
4. High nuclear grade (as per Nottingham grading system).
5. No tubule formation.

Tubular carcinoma

General

Epidemiology:

• Typically excellent prognosis.
• Hormone receptors commonly present.

Microscopic

Features:

• Well-formed tubules.
• Myoepithelial cells absent.
• +/- Cribriform spaces.
• Apocrine snouts typical.
• +/- Calcification.
• Angled ducts common: "prows" - important feature (low power).
• Looks benign to the uninitiated -- IMPORTANT.

Notes:

• Prow = front of a ship.

DDx:

• Benign sclerosing lesion.

Metaplastic carcinoma

General

• May be difficult to diagnosis.
• Prognosis - poor.

Microscopic

Features:

• Malignant mesenchymal elements:
  • Spindle cells
• +/-Adenocarcinoma.
• +/-Squamous carcinoma.

Notes:

• If it looks like squamous cell carcinoma... in the breast it is metaplastic carcinoma.

Images: Metaplastic carcinoma (breastpathology.info).

IHC

• S100 -ve (r/o melannoma).
• AE1/AE3 +ve (epithelial elements only).
• CK7 +ve (epithelial elements only).
• p63 +ve (epithelial elements only).
• Vimentin +ve.
• Desmin -ve.
• EMA -ve. (???)

Grading breast cancer

Most common system: Nottingham (aka Scarff-Bloom-Richardson) which is based on:

1. Nuclear grade.
   • Small, regular (1.5-2x RBC dia.) = 1.
   • Moderated variability = 2.
   • Marked variation (>2.5x RBC dia.) = 3.
2. Tubule formation.
   • Majority of tumour - tubules >75% = 1.
   • Moderate - 10% to 75% = 2.
   • Minimal <10% = 3.
3. Mitotic rate.
   • 0-5 mitosis/10 HPF (1.52 mm^2 --or-- 0.0152 mm^2 * 10) = 1.
   • 6-10 mitosis/10 HPF (1.52 mm^2) = 2.
   • >11 mitosis/10 HPF (1.52 mm^2) = 3.

Mnemonic: TMN = tubule formation, mitotic rate, nuclear grade.

Notes:

• Elston & Ellis devised the system that is used. They also wrote a follow-up article in 2002.

Note about mitosis counting

• One MUST adjust for the size of the field of view.

• Most of the Resident scopes have an eye piece diameter of 22 mm. Therefore, the field diameter at 40 X is approximately 22 mm / 40 X ~= 0.55 mm and the field of view is pi/4*(0.55 mm)^2 = 0.2376 mm^2.
  • Thus, on a resident scope (with a FOV of 0.2376 mm^2) one should sample 6 or 7 fields of view (FsOV).
    • Calculation: 1.52 mm^2 (sampling area) / 0.2376 mm^2 (area / FOV ) = 6.40 FsOV.

• RANT: Sampling 10 fields, where the field of view (FOV) is 0.152 mm^2, is not the same as sampling ten fields, where the FOV is 0.312 mm^2. It surprises me that Elston & Ellis ignore the fact that "10 HPFs" on different microscopes represent different sample areas and that they do not standardize the sampling area.

Calculating Nottingham score

• Grade I = 3-5 points.
• Grade II = 6-7 points.
• Grade III = 8-9 points.

Notes:

• I've found most tumours are grade II.
• The mitotic score is usually 1/3.
• The nuclear score is rarely 1/3 -- even in the tubular subtype.

Staging breast cancer

Definitions:

• Isolated tumour cells: <=0.2 mm and <200 cells.
• Micrometastasis: <=0.2 cm and ( >0.2 mm or >=200 cells ).

Tumour:

• pT1: <= 2 cm.
  • pT1mic <= 0.1 cm.
  • pT1a > 0.1 cm and <= 0.5 cm.
  • pT1b > 0.5 cm and <= 1.0 cm.
  • pT1c > 1.0 cm and <= 2.0 cm.
• pT2: > 2 cm and <= 5 cm
• pT3: > 5 cm.
• pT4: chest wall or skin involvement.

Lymph nodes:

• pN0: nil.
  • pN0(i+): <=0.2 mm and <200 cells.
• pN1: 1-3 axillary LNs or internal mammary LNs.
  • pN1mi: <=0.2 cm and ( >0.2 mm or >=200 cells ).
  • pN1a.
  • pN1b.
  • PN1c.
• pN2 4-9 positive LNs; internal mammary LNs or axillary LNs.
• pN3.

Lymphovascular invasion

There are famous criteria for lymphovascular invasion (LVI).

Rosen criteria for LVI:

1. Must be outside of the tumour proper.
   • LVI is usually very close -- typically within 0.1 cm.
2. Contour of cells should differ from possible vessel wall.
   • DCIS with retraction artifact mimicing LVI has a contour that matches its surrounding fibrous tissue.
3. Endothelium (usu. flat) should be visible.
4. Lymphatics are found adjacent to blood vessels - vessels should be present in the vicinity.

Memory device LUBE-O:

• LVI has a Unique contour, Blood vessels and Endothelium in the vicinity, and is Outside of the tumour.

Other

Paget's disease

Main article: Paget disease of the breast

General

• Associated with underlying breast carcinoma.

Notes:

• Unrelated to Paget disease of the bone.

Microscopic

Features:

• Cells in the epidermis:
  • Epitheliod morphology (round/ovoid).
  • Cells nested or single.
  • Clear/pale cytoplasm key feature - may also be eosinophilic.
  • Large nucleoli.

Images:

• Paget's disease - low mag. (WC).
• Paget's disease - high mag. (WC).

IHC & DDx:

• See Paget disease.

Sentinel lymph node biopsy

General

• Used for staging, positive LNs = poorer prognosis.

Notes:

• If there is no palpable disease, there is no mortality benefit from axillary lymph node dissection, i.e. positive axillary lymph nodes can be left in situ without affecting outcome.
  • This does not negate the fact that a positive sentinel LN biopsy (vs. negative sentinel LN biopsy) portends a poorer prognosis.

Microscopic

Features:

• Atypical cells.
  • Nuclear changes of malignancy:
    • Nuclear enlargement + variation in size.
    • Variation in shape.
    • Hyperchromasia and variation in staining.
  • Usually in the subcapsular sinuses.

Pitfalls:

• Naevus cell rests.

IHC

Some hospitals use:

• CAM5.2 (LMWK) - to look for isolated tumour cells and small lymph node metstases.

Missed macrometastases

The effect is small of missed macrometastases.

Trivia

Tumour size and lymph node metastases

There is a paper that calculates the probability of lymph node mets based on tumour size. The developed formula is:

$L_{To-Nodes}=1-exp(-Q_{n}D^{Z})$

Where:

• $L_{To-Nodes}$ = the probability of the lymph nodes being positive.
• D = the largest dimension of the tumour in millimetres.
• Z = 1.0041.
• $Q_{n}$ = 0.019.

Selected values

Tumour size (mm)	Probability
5	9 %
10	17 %
15	25 %
20	32 %
25	38 %
30	44 %
35	49 %
40	54 %
45	58 %
50	62 %

Natural history

There is a theory that up to 22% of small (radiographically detected) breast tumours regress, based on an analysis in a large population. The study is supported by NCI's SEER data. Also, it generated many comments.

See also

• Breast.
• Non-invasive breast cancer.