Estrogen re-exposure may eliminate dangerous breast clones through abnormal spindle geometry
After prolonged estrogen deprivation, re-exposure may eliminate dangerous breast clones by forcing cells with abnormal centrosomes through multipolar divisions. Death before mitosis or continued elimination after verified spindle correction would reject this mechanism.
Stage of verification
- Hypothesis published2026-10-03
- Indirect evidenceAssessed at 4 of 10
- Direct testAwaited
Map of the hypothesis
Hover over an icon or tap it to see its name.
Where in the body
Biological function
The biological function description is being prepared
Kind of knowledge gap
A double ring marks the main placement where a group contains several values.
Target map
Every target of every published hypothesis, each with the actions a hypothesis can propose on it. The targets and the actions of this hypothesis are drawn solid.

Harmful clone
Cooperating dangerous cells in breast tissue
A local population of dangerous breast cells whose descendants depend on survival signals from cooperating cells
Where this hypothesis actsAfter prolonged estrogen deprivation, in clones with centrosomal abnormalities
Hypotheses on this target 2
Clearance restoration
Elimination1
Immunosuppression
Population balance1

What is proposed
Elimination
Eliminate clones with centrosomal abnormalities through estrogen re-exposure
With whatSmall molecule
HowRe-expose estrogen-deprived cells to estrogen to drive multipolar divisions; assess irreversible lineage loss after washout and replating
Possible result
Possible irreversible loss of abnormal clones while neighboring cells with normal centrosomes continue dividing
From the recordRe-exposure forces those cells through multipolar divisions that eliminate their reproductive capacity, while neighboring cells with normal centrosomes divide successfully.

Rhythm or programme
Centrosome organization
The organization of centrosomes that provides spindle-organizing geometry during cell division
Where this hypothesis actsEstrogen-deprived breast cells undergoing estradiol re-exposure
Hypotheses on this target 1
Inhibition
Activation
Function preservation
Feedback restoration
Rhythm restoration
Direct measurement
What is proposed
Correct centrosome number or restore bipolar spindle assembly
With whatControlled genetic model
HowUse inducible centrosome-manipulation systems while preserving comparable cell-cycle entry, intracellular estradiol and receptor activation
Possible result
Expected preservation of long-term clonogenic survival despite estrogen re-exposure
From the recordCorrecting centrosome number or restoring bipolar spindle assembly preserves long-term clonogenic survival despite unchanged intracellular estradiol and receptor activation.
All targets of the lab
Every target read from the published hypotheses, each kind around its pictogram. A larger mark means more hypotheses act on that target. Point at a mark and the actions proposed on it branch out of it.
Solid and named: the targets of this hypothesis
Explore in depth
The logic
The train of thought that ends in this hypothesis. Each stage is the reason the next exists. The master question narrows to a goal, the goal to an unknown nobody has closed, the unknown to the hypothesis proposed here. Every step below says what it rests on and what carries it.
A return of estrogen after a long absence might destroy some potentially dangerous breast cells while helping others multiply. The unexpected proposal is that the difference lies in the physical arrangement of the machinery that separates a dividing cell’s genetic material. This is a hypothesis generated by the pipeline, not a measured explanation of how menopause could inform radical lifespan extension.
- Prolonged estrogen deprivation permits extra or disorganized centrosomes to persist in slowly dividing breast cells.
- The abnormal division machinery remains present while the cells divide infrequently.
- Estrogen re-exposure pushes susceptible cells from infrequent division into divisions organized around several poles instead of two.
- Those abnormal divisions are proposed to prevent the affected cell families from producing lasting descendants.
- Neighboring cells with normal centrosomes are predicted to divide successfully under the same exposure.
A sorting machine with its guides pointing in conflicting directions may sit idle without revealing the fault. Restarting it exposes the faulty arrangement as material is pulled toward several exits.
Where the picture breaks: The picture explains why restarting activity could reveal a stored physical defect. It does not establish that abnormal cell division permanently eliminates descendants, or that estrogen acts only through restarting division.
- Master questionstep 01 of 04
Changes associated with menopause, the end of menstrual cycles, might provide knowledge useful for radically extending life.
Rests on: The goal treats menopause-associated changes as a possible source of knowledge about lifespan extension.
AssumptionThe usefulness of those changes for radical lifespan extension is taken as a research premise; the supplied material does not establish that connection.
- Goal pillarstep 02 of 04
An adjustment that helps compensate for a biological change may also cause harm that appears later, and those effects need to be separated.
Rests on: The lifespan-extension goal requires identifying which consequences of menopause-associated changes would help or harm that goal.
AssumptionThe pillar assumes that compensatory benefit and delayed collateral harm provide a useful way to investigate the master question. The master question itself does not supply this framework.
- Gap questionstep 03 of 04
Prolonged estrogen deprivation might change later hormone exposure from favoring dangerous breast clones, families of cells descended from a common starting cell, to eliminating them. Comparing matched exposures is intended to separate this history effect from differences in hormone preparation, detection and participant selection.
Rests on: The preceding pillar calls for distinguishing benefit from harm; this question makes breast cell expansion versus elimination the concrete distinction.
AssumptionBreast clone behavior is selected as a relevant instance of the pillar’s benefit–harm problem. Its contribution to radical lifespan extension, and what qualifies a clone as dangerous, are not established in the supplied material.
- Hypothesisstep 04 of 04
Long estrogen deprivation is proposed to leave susceptible cells with extra or disorganized centrosomes, structures that help organize cell division. When estrogen returns, those cells are predicted to divide toward several poles instead of two and lose the ability to produce lasting descendants, while neighboring cells with normal division machinery continue multiplying.
Rests on: The preceding question explicitly supplies the possibility of a history-dependent switch from expansion to elimination at matched exposure. The endpoint supplies a proposed physical explanation and predictions that distinguish it from competing explanations.
Stated in the chain
What is carried, and what is not. None of the six screened sources establishes the proposed links from persistent centrosome abnormalities to abnormal division and lasting clone elimination: S2, an abstract from The Journal of Steroid Biochemistry and Molecular Biology in 2001, reports that estrogen re-exposure reduced activity of aromatase, an enzyme involved in estrogen production, in previously deprived breast cancer cells, but does not test those links. S4, a 2023 Clinical Cancer Research paper, reports estrogen-triggered damage to genetic material that depends on the estrogen receptor, a protein through which cells respond to estrogen, and is enhanced by adaptation to hormone deprivation; this supplies a competing route to injury, without testing whether centrosome abnormalities determine elimination or establishing the proposed sequence end to end.S2S4
Where the reasoning is carried by something unstated · 3
- Master question. The usefulness of those changes for radical lifespan extension is taken as a research premise; the supplied material does not establish that connection.
- Goal pillar. The pillar assumes that compensatory benefit and delayed collateral harm provide a useful way to investigate the master question. The master question itself does not supply this framework.
- Gap question. Breast clone behavior is selected as a relevant instance of the pillar’s benefit–harm problem. Its contribution to radical lifespan extension, and what qualifies a clone as dangerous, are not established in the supplied material.
How a result here could mislead · 3
- Correcting division machinery could appear to rescue the cells simply because the manipulation stops them entering division, removing the occasion for injury. What closes it: The specified rescue comparison must verify both correction of the abnormal machinery and comparable entry into division. Intracellular estradiol, the form of estrogen measured inside the cells, and activation of its receptor must also remain comparable.
- A temporary pause in growth, or failure to see descendants during imaging, could be mistaken for irreversible elimination of a cell family. What closes it: The design requires counting viable descendants after washout, removal of the treatment, and replating, transfer into fresh culture conditions. The follow-up period and criterion for lasting reproductive loss must be fixed in advance; neither is specified in the supplied material.
- Abnormal divisions could accompany elimination without causing it. A centrosome manipulation might also change how many neighboring cells remain, making apparent rescue compatible with the rival explanation based on cooperation between cells. What closes it: Direct imaging must establish the order of abnormal division and lineage loss, meaning loss of a cell family’s ability to continue producing descendants. The correction and defect-introduction comparisons must also measure or control local cell density and cooperative support; the supplied design does not specify that comparison.
What would make this wrong. The proposed explanation would fail if susceptible cell families continued to lose lasting reproductive capacity after verified correction of centrosome number or division organization, with comparable entry into division, intracellular estradiol and receptor activation. Irreversible loss occurring before division would also contradict the proposed route. Either observation would weaken this explanation without by itself establishing which rival mechanism caused the loss.
What it would change. If the proposed mechanism held, deprivation history alone would be insufficient to predict whether renewed estrogen exposure expands or eliminates dangerous breast cell families; the condition of their division machinery would also matter. Work connecting menopause to lifespan extension would then need to distinguish removal of susceptible families from expansion of surviving ones. Even a successful test in laboratory cultures would not establish selective protection in people, reduced breast cancer risk, or radical lifespan extension.
Sources read · 6
Resistance to everolimus driven by epigenetic regulation of MYC in ER+ breast cancers. · Oncotarget · 2015
“Furthermore, LTED derivatives of MCF7 and ZR75 both resulted in significant loss of ER expression”
Does not settle: This source does not test estrogen re-exposure, centrosome amplification or organization, spindle geometry, multipolar division, mitotic catastrophe, clone elimination, matched intracellular steroid exposure, or whether susceptibility is restricted to clones with centrosomal abnormalities.
The potential role of estrogen in aromatase regulation in the breast. · The Journal of steroid biochemistry and molecular biology · 2001
“Re-exposure of LTED cells to estrogen reduced aromatase activity to the levels of the wild-type MCF-7 cells.”
Does not settle: The source does not establish clone elimination, mitotic catastrophe, centrosome amplification or organization, multipolar spindle geometry, reproductive capacity, matched intracellular steroid exposure, or which clones would be susceptible.
Nucleosome destabilization by nuclear non-coding RNAs. · Communications biology · 2020
“The ESR1 gene, encoding the estrogen receptor-α, ER, is upregulated in ER-positive breast cancer cells when they undergo adaptation to the hormone-depleted environment, defined as long-term estrogen deprivation (LTED)”
Does not settle: This source does not examine estrogen re-exposure, centrosome amplification or organization, spindle geometry, multipolar division, mitotic catastrophe, clone elimination, reproductive capacity, matched intracellular steroid exposure, or whether susceptibility is independent of receptor renewal, transcriptional memory, or unrepaired DNA lesions.
Estrogen Therapy Induces Receptor-Dependent DNA Damage Enhanced by PARP Inhibition in ER+ Breast Cancer. · Clinical cancer research : an official journal of the American Association for Cancer Research · 2023
“Herein, we demonstrate that E2 induces ER-dependent S-phase-specific DNA damage and R-loop accumulation that is exacerbated by ER overexpression and adaptation to growth in hormone-depleted conditions.”
Does not settle: The source does not test centrosome amplification or organization, spindle geometry, multipolar division, mitotic catastrophe, reproductive elimination, matched intracellular steroid exposure, or whether susceptibility is restricted to clones with centrosomal abnormalities.
Unraveling Vulnerabilities in Endocrine Therapy-Resistant HER2+/ER+ Breast Cancer. · Endocrinology · 2023
“To mimic ETR to aromatase inhibitors (AIs), we developed 2 long-term estrogen deprivation (LTED) cell lines from BT-474 (BT474) and MDA-MB-361 (MM361).”
Does not settle: The abstract does not test estrogen re-exposure, mitotic catastrophe, centrosome amplification or organization, spindle geometry, multipolar division, clone-selective reproductive elimination, or matched intracellular steroid exposure. It therefore does not establish whether centrosomal abnormalities determine which estrogen-deprived breast cancer clones are eliminated rather than expanded.
Life following aromatase inhibitors--where now for endocrine sequencing? · Breast cancer research and treatment · 2005
“Data from in vitro models have suggested that acquired AI resistance is due to enhanced sensitization to low estrogen levels during long-term estrogen deprivation (LTED).”
Does not settle: The abstract does not examine estrogen re-exposure, mitotic catastrophe, spindle geometry, centrosome amplification or organization, clone elimination, matched intracellular steroid exposure, or whether susceptibility is restricted to clones with centrosomal abnormalities.
The gap this hypothesis explains
Two established results predict opposite outcomes, and both cannot be right.
Does prolonged estrogen withdrawal switch dangerous breast cell groups from expansion to elimination when later exposure is identical?
Original wording · exactly as the pipeline generated it
Does prolonged estrogen deprivation reverse subsequent exposure from selecting dangerous breast clones to eliminating them, and can matched-exposure experiments distinguish this history effect from formulation, detection and participant-selection effects?
What this question is asking
The question concerns whether a long period with little estrogen changes how potentially cancerous groups of breast cells respond when estrogen returns. It asks whether later exposure favors those groups or eliminates them, comparing different exposure histories while holding the later hormone exposure equal. It also asks whether such comparisons can separate an effect of history from differences in hormone preparation, how disease is detected, and which participants enter a study. The question assumes that estrogen can initially favor dangerous groups and draws on reports that some previously deprived breast cancer cells instead die after estrogen returns; whether these observations establish a reversal in people is part of what needs auditing.
- Estrogen and estradiol
- Estrogen names a class of hormones; estradiol is a particular member used in several supplied studies. In these sources, estrogen-related responses include both cell multiplication and programmed cell death, depending on the cellular setting.
- Estrogen deprivation or withdrawal
- A period with reduced estrogen availability. Its duration and biological setting can vary, so laboratory deprivation, hormone-blocking treatment, and menopause are not established here as interchangeable conditions.
- Breast clone and clonal selection
- A clone is a group of breast cells descended from a common starting cell. Selection means that conditions favor some groups' survival or multiplication relative to others; it does not necessarily mean creating new harmful cells.
- Dangerous breast cell group
- The question's label for cells capable of contributing to cancer. The supplied material gives no shared test or threshold defining which groups qualify as dangerous.
- Proliferation
- An increase in cell number through cell division. It is the growth response that the question contrasts with cell death.
- Apoptosis
- A regulated process through which a cell dies. Increased apoptosis in a cell population does not by itself demonstrate elimination of an entire dangerous clone.
- Hormone-blocking treatment and acquired resistance
- Hormone-blocking treatment reduces hormone production or interferes with hormone action. Acquired resistance means cancer cells become less responsive to that treatment over time; the supplied sources describe some resistant cells as vulnerable to renewed estrogen exposure.
- Estrogen receptor
- A cellular protein through which estrogen can influence cell activity. S6 places the receptor in the nucleus, the compartment containing genetic material, at the start of the stress response associated with estrogen-induced death.
- Cellular stress response and inflammation
- A cellular stress response is a change in cell activity when normal functioning is challenged. Inflammation is a tissue response to injury or disturbance; S1 refers to proteins associated with these processes when describing the change in estrogen response.
- MCF-7 and cell models
- MCF-7 is the name of a breast cancer cell line grown for laboratory research. Related groups derived from it can develop different responses, and findings in those cells do not by themselves establish responses in normal breast tissue.
- Menopause and menopausal status
- Menopause is the life transition when menstrual cycles permanently end. Menopausal status describes a person's position relative to that transition; S5 identifies it as relevant to successful estrogen treatment.
- Matched exposure and hormone history
- Matched exposure means holding later hormone treatment conditions equal in the comparison. Hormone history means earlier exposure, deprivation, or treatment; separating people into history groups does not by itself establish that history caused a difference.
- Formulation, detection, and participant-selection effects
- Formulation effects arise from differences in the hormone preparation being compared. Detection effects arise from differences in finding disease, while participant-selection effects arise from differences in who enters a study; each could affect observed outcomes without establishing a causal effect of prior deprivation.
- Follow-up and clinical outcomes
- Follow-up is observation over time after an initial treatment or measurement. Clinical outcomes concern what happens in people, such as later cancer occurrence, rather than only changes in laboratory cell growth or death.
Estrogen exposure initially selects dangerous breast clones, while prolonged estrogen deprivation can change the subsequent response from growth to apoptosis.
Estrogen is a hormone, and a breast clone is a group of cells descended from one starting cell. The assumption is that estrogen first favors potentially harmful groups, but a long period without it changes those cells so that renewed exposure makes them die. If established in the relevant breast tissue, this would make prior exposure history a possible explanation for opposite responses to later estrogen.
S1 reports a switch from growth to programmed cell death after long-term treatment that blocks hormone action, and S7 and S8 report death responses in estrogen-deprived breast cancer cell models. S5 reports that successful estrogen treatment depends on withdrawal duration and menopausal status, but only its abstract was supplied. These findings support the narrower claim that certain adapted breast cancer cells can die after estrogen exposure. They do not establish an initial phase of dangerous-clone selection, a causal reversal of that selection in patients, or the same response in normal breast tissue after menopause. The supplied sources also do not substantiate the pipeline's assertion that clinical estimates differ by study design.S1S5S7S8
The same question asked without the part nothing read establishes:
- With later estrogen exposure held equal, how does prior deprivation affect the survival and expansion of potentially dangerous breast cell groups?
- Do differences associated with prior estrogen deprivation persist when hormone preparation, disease detection, and participant selection are accounted for?
- History reverses expansion into elimination If established, prolonged deprivation would change the cells' response so that otherwise growth-favoring estrogen exposure instead removes the dangerous groups. Exposure history would then change the direction of the response, although lasting protection would still require evidence that those groups remain absent during exposure and later follow-up.
- History does not reverse the response If dangerous groups continue to survive or expand after deprivation, death responses in selected laboratory cells would not establish elimination in the breast tissue at issue. Prior withdrawal could not then be treated as evidence that later estrogen exposure changes from harmful selection to protection.
- Responses differ among cell groups or over time If estrogen kills some groups while others survive or expand, an early reduction in cell growth could coexist with persistence of dangerous cells. In that case, early benefit would not by itself establish the direction of harm over subsequent years.
If estrogen favors the survival or multiplication of dangerous breast cell groups, those groups could become more prominent during exposure. If previous deprivation instead makes those cells die when estrogen returns, the same later exposure could have a different consequence. However, death among some cells does not establish elimination of every dangerous group or protection over subsequent years. Treating a short-term response as proof of lasting protection could therefore mischaracterize delayed harm, while assuming that estrogen always promotes growth could miss the death response reported in particular breast cancer models.
RL-2 clinical estimates differ by design; RL-1 deprivation models predict apoptosis rather than growth, without validation in normal menopausal breast tissue.
Within each endocrine-history stratum, early benefit must remain separated from proliferative harm throughout exposure and years of follow-up.
Whether deprivation causally reverses clonal selection remains unknown, preventing determination of whether exposure history changes the direction of delayed harm.
The mechanism it proposes
The engine's own statement of the hypothesis, in full.
HERETICAL: Prolonged estrogen deprivation makes dangerous breast clones vulnerable to estrogen-driven mitotic catastrophe rather than converting estrogen into a direct apoptotic signal. Deprivation permits centrosome amplification or abnormal centrosome organization to persist in slowly cycling cells. Re-exposure forces those cells through multipolar divisions that eliminate their reproductive capacity, while neighboring cells with normal centrosomes divide successfully. The stored susceptibility is physical spindle-organizing geometry, not receptor renewal, transcriptional memory or unrepaired DNA lesions. This predicts genuine elimination at matched intracellular steroid exposure, but only in clones with the relevant centrosomal abnormality. Establishing that boundary would help stabilize SPV_7 by distinguishing exposures that eliminate such clones from exposures that expand surviving dangerous lineages.
Testing and possible results
The prediction that would tell it apart
A hypothesis that predicts what its rivals predict is not worth running an experiment over. This is the observation on which this one differs.
In sister cultures randomized to different deprivation durations, irreversible lineage loss occurs predominantly after directly observed multipolar mitoses. Correcting centrosome number or restoring bipolar spindle assembly preserves long-term clonogenic survival despite unchanged intracellular estradiol and receptor activation. Conversely, introducing the corresponding centrosomal abnormality into short-deprived cells reproduces susceptibility. Death before mitosis, or continued elimination after verified spindle correction, rejects this explanation in favor of the ribosomal or lysosomal rivals.
Would tell it apart from at least one rival. The prediction specifies observable temporal relationships, intervention outcomes, and explicit rejection conditions. No rival prediction is supplied for comparison. A paper already fetched for this hypothesis bears on it.
What testing it would take
The engine's own read on whether this is testable with methods that already exist.
Organoid lineage imaging, centrin and spindle reporters, and inducible centrosome-manipulation systems permit an experimental test. Manipulations must preserve comparable cell-cycle entry: simply arresting proliferation would not establish spindle-specific rescue. Count viable descendants after washout and replating rather than interpreting temporary growth arrest as elimination.
Other explanations
Every other hypothesis the engine wrote for the same gap, and the observation that would separate the two.
In sister cultures randomized to different deprivation durations, irreversible lineage loss occurs predominantly after directly observed multipolar mitoses. Correcting centrosome number or restoring bipolar spindle assembly preserves long-term clonogenic survival despite unchanged intracellular estradiol and receptor activation. Conversely, introducing the corresponding centrosomal abnormality into short-deprived cells reproduces susceptibility. Death before mitosis, or continued elimination after verified spindle correction, rejects this explanation in favor of the ribosomal or lysosomal rivals.
- Rival 01 of 03What would separate them
Estrogen may eliminate dangerous breast clones when cooperative support falls below a threshold predicts: At identical intracellular estradiol exposure and total epithelial density, changing the local abundance of cooperating dangerous cells produces a reproducible sign change in their absolute net growth. Reconstituting cooperating cells, or their experimentally verified survival signal, rescues long-deprived lineages without altering their centrosomes, ribosomal collision response or lysosomal integrity. Diluting the same cooperating population below its fitted threshold makes short-deprived lineages decline. Failure of reciprocal population reconstruction to transfer the response rejects this IH in favor of a cell-intrinsic mechanism.
- Rival 02 of 03What would separate them
Estrogen re-exposure may kill deprived breast cell lineages through ribosome collisions predicts: Deprived lineages show a re-exposure-specific increase in collided-ribosome footprints followed by ZAKα–p38/JNK activation before their first mitosis or lysosomal leakage. ZAKα disruption preserves clonogenic survival despite persistent collision footprints and matched estrogen signaling; wild-type ZAKα restores killing, whereas a collision-sensing-defective construct does not. Equalizing cooperative-cell density does not rescue. Absence of the predicted collision sequence, or failure of selective ZAKα disruption to rescue, favors the other IHs.
- Rival 03 of 03What would separate them
Estrogen re-exposure may kill breast cell lineages by rupturing deprivation-altered lysosomes predicts: During matched re-exposure, lysosomal leakage and cytosolic cathepsin activity precede mitochondrial permeabilization, caspase activation and the first mitosis. Independently validated suppression of the responsible cathepsins or prevention of lysosomal rupture restores post-washout clonogenic survival, whereas ZAKα disruption, spindle correction and cooperative-population reconstruction do not. Executioner-caspase inhibition alone fails to preserve the lineage. Leakage only after another death pathway activates, or failure of lysosome-specific rescue, rejects this initiating mechanism.
Why this is not the mainstream account
The engine is asked to say what its hypothesis would overturn and what would surprise a specialist. This is its answer.
Estrogen caused centrosome amplification before overt mammary carcinoma in ACI rats, establishing a potentially dangerous mitotic substrate rather than demonstrating the proposed elimination mechanism. [Primary estrogen–centrosome study](https://pmc.ncbi.nlm.nih.gov/articles/PMC539804/). Separately, physiologic estradiol eliminated long-deprived MCF-7:5C tumors through an experimentally implicated mitochondrial death pathway. [Primary deprivation study](https://pubmed.ncbi.nlm.nih.gov/16333030/). The proposed link between these observations remains untested.
Breast endocrine oncology; the textbook chapter topic 'Acquired endocrine resistance and estrogen-induced apoptosis' would require revision if deprivation-dependent elimination were primarily a spindle-geometry catastrophe and canonical apoptosis were a downstream consequence.
Repairing spindle geometry converts estradiol-induced eradication into sustained expansion of the same deprived dangerous lineage without reducing estrogen signaling; creating the geometry defect confers susceptibility without prolonged deprivation.
A targeted literature search identified established estrogen-induced apoptotic models and estrogen-associated centrosome amplification, but did not identify this specific deprivation-to-spindle-catastrophe explanation. This is provisional novelty evidence, not proof that no review or perspective has proposed it.
What stands behind it
Which of the figures above have a study behind them, which are the engine's own, and what it would take to refute the hypothesis. This audit never judges the idea.
This hypothesis states no figure and cites no study, so there is nothing here to trace.
What it would take to refute it. 1 paper(s) already retrieved for this hypothesis carry its prediction’s terms. Reading them comes before running anything. Already retrieved: Fasting: How to Guide..
1 paper retrieved around this hypothesis
- Fasting: How to Guide.PMID 34067055 · full_text · 86,055 characters stored
0 citation handles extracted; 1 Europe PMC search run; 1 records examined; 1 sources stored for enrichment, 1 with full text. A citation that did not resolve is a bibliographic failure, not proof that no such paper exists, and no hypothesis is blocked by this audit.