Part II · Normal biology, carcinogenesis, and inherited susceptibility · Chapter 4

Normal breast development and physiology

The normal gland supplies the cell of origin, the timing of risk, and the stroma a tumour will use.

1 · Anatomy and cellular organization

The functional unit is the terminal duct lobular unit. A terminal duct ends in a cluster of acini, and almost all carcinomas arise from this structure rather than from the large ducts.

Every duct and acinus is lined by two layers. An inner luminal layer faces the lumen and carries the secretory and hormone-sensing function. An outer myoepithelial layer lies between the luminal cells and the basement membrane. The myoepithelium contracts during lactation. It also polarises the luminal layer and physically restrains it.

That second role is the one the rest of this book depends on. Myoepithelium plus basement membrane is the boundary whose breach defines invasion. In situ and invasive disease are separated by a structure rather than by a molecular event, which is why Breast carcinogenesis and precursor states returns to what happens to that structure.

Around the epithelium sits stroma, meaning fibroblasts, collagen, vessels, adipocytes and resident immune cells. This is not inert packing. A single-cell and spatial atlas of the adult human breast profiled 714,331 cells from 126 women and resolved 12 major cell types and 58 biological cell states, including abundant perivascular, endothelial and immune populations1. The same work found molecular differences between ductal and lobular regions.

The sampling consequence follows directly. A core of histologically normal breast returns a mixture whose composition varies with site, age and menstrual phase. Cellular architecture of the tumor microenvironment treats the same cell types once a tumour is present.

2 · Mammary stem cells, lineage hierarchy, and candidate cells of origin

Two experimental systems give different answers about the same tissue. The disagreement is methodological rather than biological, and reading it as biological has caused a great deal of confusion.

Transplantation asks what a dissociated cell can do. Small numbers of transplanted basal cells regenerate a functioning gland, which supports a multipotent stem cell at the apex of a hierarchy.

Lineage tracing asks what a cell actually does in place. Genetic tracing in the unperturbed mouse gland found long-lived unipotent stem cells within the luminal lineage and within the myoepithelial lineage2. Each renewed itself through morphogenesis, adult life and repeated cycles of pregnancy.

Both results are correct about their own question. Dissociation and transplantation reveal potential. Tracing reveals fate. A hierarchy drawn from transplantation is a map of what cells could become if the tissue were destroyed and rebuilt.

The markers used to isolate these populations carry the same caveat. Sorting for CD44 high and CD24 low enriches for tumour-initiating cells3. Aldehyde dehydrogenase activity enriches for normal and malignant mammary stem cells and carries prognostic weight4. Each defines a fraction, and a fraction is not a cell type.

One cell-of-origin question does have a firm answer. Breast tissue from BRCA1 mutation carriers harbours an expanded luminal progenitor population with factor-independent growth in culture5. Basal-like tumours in that study resembled luminal progenitors more closely than they resembled the stem-cell-enriched fraction. Basal-like is therefore a transcriptional label rather than a claim about a basal cell of origin. Subtype-specific origins are taken up in Breast carcinogenesis and precursor states.

3 · Endocrine regulation across puberty, cycling, and menopause

At puberty oestrogen drives ductal elongation and terminal end bud formation. Progesterone drives side branching, and in pregnancy it drives alveolar development.

The mechanism is not the one the clinical mental model implies. In normal epithelium the proliferating cell is usually not the oestrogen receptor positive cell. Mammary epithelium lacking oestrogen receptor alpha proliferates and contributes to all aspects of gland development when it lies close to receptor-positive cells6. Normal oestrogen action on proliferation is paracrine.

This matters for how endocrine therapy is understood. In most hormone receptor positive tumours the receptor-positive cell is itself the proliferating cell, so the dependency has become cell-autonomous. That rewiring between normal tissue and tumour is developed in Estrogen receptor signaling and endocrine resistance.

The stem compartment is governed by the same hormones without expressing their receptors. Mouse mammary stem cells lack oestrogen and progesterone receptors, yet ovariectomy reduced their number and outgrowth potential, and oestrogen with progesterone increased it7. Three weeks of letrozole was sufficient to shrink the pool in mice. RANK ligand is the proposed paracrine intermediary.

Circulating hormone levels carry risk into the postmenopausal decades. A reanalysis of nine prospective studies covered 663 women who developed breast cancer and 1,765 who did not. The relative risk in the highest quintile of serum oestradiol against the lowest was 2.00 (95% CI 1.47 to 2.71)8. Androgens tracked in the same direction.

Proliferation is not constant across the menstrual cycle. It rises in the luteal phase, when progesterone is high. A proliferation index measured on a premenopausal core therefore carries a timing term that the report does not state.

After menopause the cyclical stimulus is withdrawn. Lobules involute and are replaced by fat and collagen, which is the substrate for both Breast density, what it is and why it predicts and Aging and the changing normal microenvironment.

4 · Pregnancy, lactation, and postpartum involution

Pregnancy completes the developmental programme. Alveoli form, luminal cells differentiate into secretory cells, and the gland reaches its only fully differentiated state.

Weaning reverses it. Involution runs in two phases. The first is reversible and driven by local signals from milk stasis. The second is irreversible. It removes the secretory epithelium by apoptosis and remodels the stroma back towards its pre-pregnant configuration.

The second phase is where the oncology sits. It is a tissue remodelling programme with the features of wound healing: matrix degradation, then fibrillar collagen deposition, cyclooxygenase-2 induction, and recruitment of macrophages with an alternatively activated phenotype.

The experimental demonstration is direct. Human breast cancer cells placed in the involuting mouse mammary gland formed large tumours rich in fibrillar collagen, with high COX-2 expression and an invasive phenotype9. Inhibiting COX-2 reduced collagen fibrillogenesis, tumour growth, and infiltration of tumour cells to the lung.

The matrix changes are not incidental to that result. Collagen crosslinking stiffens the matrix and enhances integrin signalling10. Aligned collagen fibres carry prognostic weight in human breast carcinoma11.

The stem compartment moves as well. Mammary stem cell number rose transiently about eleven-fold during pregnancy in mice7.

Involution therefore supplies two of the things a new tumour needs. It follows an expansion of a susceptible population, and it builds a permissive stroma. This returns in Young adults, fertility, and pregnancy as the biology of postpartum breast cancer. In a cohort of 701 women aged 45 or younger, diagnosis within ten years of childbirth was associated with higher risk of distant metastasis12.

5 · Parity and the biology of transient and long-term risk

Parity is described as protective, and over a long enough horizon it is. Over a shorter horizon it is the opposite. Both statements come from the same data.

A pooled analysis of individual-level data from 15 prospective cohorts measured the shape of the effect, covering 18,826 incident breast cancers in women younger than 55 across 9.6 million person-years13. Compared with nulliparous women, the hazard ratio for parous women peaked about 5 years after birth at 1.80 (95% CI 1.63 to 1.99). It fell to 0.77 (95% CI 0.67 to 0.88) more than 34 years after birth. The crossover from increased to decreased risk occurred about 24 years after birth.

Three features of that analysis matter more than the headline. The pattern was driven by oestrogen receptor positive disease, and no crossover was seen for receptor-negative disease. The postpartum increase was more pronounced in women with a family history. It was greater in women who were older at first birth and in those with more births.

Breastfeeding acts separately and in one direction only. A collaborative reanalysis pooled 47 epidemiological studies from 30 countries, covering 50,302 women with breast cancer and 96,973 without. Relative risk fell by 4.3% for every 12 months of breastfeeding (95% CI 2.9 to 5.8)14. That was in addition to a fall of 7.0% for each birth (95% CI 5.0 to 9.0).

The two limbs have different biology. The early limb is involution, described in Pregnancy, lactation, and postpartum involution, acting on a recently expanded progenitor pool. The late limb is differentiation, which permanently reduces the population available for transformation.

Caution

Parity recorded as ever or never averages an early increase against a late decrease. For a woman who gave birth three years ago, recent childbirth is a risk factor and not a protective one13. A risk model that treats parity as a binary variable cannot represent this, and Risk assessment and prevention returns to what such models can and cannot do.

6 · Breast density, what it is and why it predicts

Density is a measurement before it is a biological property. Mammographic density is the proportion of the imaged breast that attenuates X-rays, which is fibroglandular tissue against fat. It reports epithelium and stroma together and does not separate them.

Two different things follow from a dense mammogram, and they should be kept apart.

The first is masking. Cancer and dense tissue attenuate similarly, so a lesion is harder to see.

The second is risk that is independent of detection. Three nested case-control studies contributed 1,112 matched case-control pairs. Women with density in 75% or more of the mammogram had an odds ratio for breast cancer of 4.7 against women with density below 10% (95% CI 3.0 to 7.4)15. For screen-detected cancers the odds ratio was 3.5 (95% CI 2.0 to 6.2). For cancers found less than 12 months after a negative screening examination it was 17.8 (95% CI 4.8 to 65.9).

The gap between those last two figures is masking made visible. The elevated risk persisted for at least 8 years after study entry, which masking alone does not explain.

The biology is stromal as much as epithelial. Dense tissue is collagen-rich, and collagen crosslinking stiffens the matrix and amplifies integrin signalling10. Fibre alignment itself carries prognostic information11.

Density falls with age and after menopause. Categorical reporting rests on reader judgement, so a change of recorded category between screening rounds may be a change in the reader rather than in the breast. Risk modelling and supplemental imaging policy follow in Breast density and other risk modifiers and Dense breasts, supplemental imaging, and notification policy.

7 · Aging and the changing normal microenvironment

Age is the largest risk factor for breast cancer, and the mechanism sits in the tissue rather than in cumulative exposure alone. Three changes are separable.

The first is clonal. Histologically normal epithelium accumulates somatic mutations and clonal expansions across a lifetime. Phylogenetic analysis of microdissected cancer and non-cancer lesions dated acquisition of the der(1;16) chromosome, found in roughly a fifth of breast cancers, to the interval between puberty and late adolescence16. A common ancestor of cancer and non-cancer clones emerged by the patient's early thirties. Those clones then replaced surrounding epithelium and occupied a large area of premenopausal breast tissue by the time cancer was diagnosed. Field effect and clonal expansion in histologically normal tissue develops what this means for the field effect.

The second change is architectural, and it runs in the protective direction. Lobules involute with age, and failure to involute carries risk. A benign breast disease cohort of 8,736 women was followed for a median of 17 years, with relative risk measured against a population registry17. Relative risk was 1.88 with no lobular involution (95% CI 1.59 to 2.21), 1.47 with partial involution (95% CI 1.33 to 1.61) and 0.91 with complete involution (95% CI 0.75 to 1.10). Among the women with atypia the corresponding relative risks were 7.79, 4.06 and 1.49. The aging that fails to happen is the risk.

The third change is stromal and immune. Aged stroma alters matrix composition and secretory phenotype, and in melanoma such changes can release disseminated cells from dormancy18. Dormancy, residual disease, and late recurrence develops dormancy directly.

One measurement consequence belongs here. Clonal haematopoiesis becomes common with age19, and its variants appear in plasma cell-free DNA20. A mutation detected in blood from an older patient is not automatically a tumour mutation.

References

All 20 references
  1. Kumar T, Nee K, Wei R, et al. A spatially resolved single-cell genomic atlas of the adult human breast. Nature 2023 620:181-191. PMID 37380767
  2. Van Keymeulen A, Rocha AS, Ousset M, et al. Distinct stem cells contribute to mammary gland development and maintenance. Nature 2011 479:189-193. PMID 21983963
  3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003 100:3983-3988. PMID 12629218
  4. Ginestier C, Hur MH, Charafe-Jauffret E, et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 2007 1:555-567. PMID 18371393
  5. Lim E, Vaillant F, Wu D, et al. Aberrant luminal progenitors as the candidate target population for basal tumor development in BRCA1 mutation carriers. Nat Med 2009 15:907-913. PMID 19648928
  6. Mallepell S, Krust A, Chambon P, Brisken C. Paracrine signaling through the epithelial estrogen receptor alpha is required for proliferation and morphogenesis in the mammary gland. Proc Natl Acad Sci U S A 2006 103:2196-2201. PMID 16452162
  7. Asselin-Labat ML, Vaillant F, Sheridan JM, et al. Control of mammary stem cell function by steroid hormone signalling. Nature 2010 465:798-802. PMID 20383121
  8. Key T, Appleby P, Barnes I, Reeves G. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst 2002 94:606-616. PMID 11959894
  9. Lyons TR, O'Brien J, Borges VF, et al. Postpartum mammary gland involution drives progression of ductal carcinoma in situ through collagen and COX-2. Nat Med 2011 17:1109-1115. PMID 21822285
  10. Levental KR, Yu H, Kass L, et al. Matrix crosslinking forces tumor progression by enhancing integrin signaling. Cell 2009 139:891-906. PMID 19931152
  11. Conklin MW, Eickhoff JC, Riching KM, et al. Aligned collagen is a prognostic signature for survival in human breast carcinoma. Am J Pathol 2011 178:1221-1232. PMID 21356373
  12. Goddard ET, Bassale S, Schedin T, et al. Association between postpartum breast cancer diagnosis and metastasis and the clinical features underlying risk. JAMA Netw Open 2019 2:e186997. PMID 30646210
  13. Nichols HB, Schoemaker MJ, Cai J, et al. Breast cancer risk after recent childbirth: a pooled analysis of 15 prospective studies. Ann Intern Med 2019 170:22-30. PMID 30534999
  14. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and breastfeeding: collaborative reanalysis of individual data from 47 epidemiological studies in 30 countries, including 50302 women with breast cancer and 96973 women without the disease. Lancet 2002 360:187-195. PMID 12133652
  15. Boyd NF, Guo H, Martin LJ, et al. Mammographic density and the risk and detection of breast cancer. N Engl J Med 2007 356:227-236. PMID 17229950
  16. Nishimura T, Kakiuchi N, Yoshida K, et al. Evolutionary histories of breast cancer and related clones. Nature 2023 620:607-614. PMID 37495687
  17. Milanese TR, Hartmann LC, Sellers TA, et al. Age-related lobular involution and risk of breast cancer. J Natl Cancer Inst 2006 98:1600-1607. PMID 17105983
  18. Fane ME, Chhabra Y, Alicea GM, et al. Stromal changes in the aged lung induce an emergence from melanoma dormancy. Nature 2022 606:396-405. PMID 35650435
  19. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med 2014 371:2488-2498. PMID 25426837
  20. Razavi P, Li BT, Brown DN, et al. High-intensity sequencing reveals the sources of plasma circulating cell-free DNA variants. Nat Med 2019 25:1928-1937. PMID 31768066