🧬 Aging Wiki

parp-inhibitors

Properties1
aliasesPARPi, PARP inhibitors class, olaparib class, niraparib class

15 min read

PARP inhibitors

PARP inhibitors (PARPi) are competitive NAD+-mimetic small molecules that block parp1’s catalytic poly(ADP-ribosyl)ation activity and, for the more potent agents, physically trap the PARP1–DNA complex at single-strand break sites. The class is FDA-approved for oncology (BRCA1/2-mutant cancers, leveraging synthetic lethality with homologous recombination deficiency) and represents one of the most successful examples of mechanism-based precision oncology. The aging-relevance hypothesis is distinct and more speculative: chronic PARP1 hyperactivation in aged cells — driven by accumulated DNA damage burden — depletes cellular NAD+, potentially impairing sirtuin activity and mitochondrial function. PARP inhibitors might partially restore this NAD+ balance, but they also carry a documented risk of increasing clonal hematopoiesis, which itself accelerates cardiovascular and hematological aging risk. This page documents both the oncology-established biology and the emerging, contested aging axis.

This is a class page. Individual compound pages when seeded: olaparib, rucaparib, niraparib, talazoparib, veliparib. Mechanistic pathway anchors: base-excision-repair, homologous-recombination, parp1.

Class composition

AgentApprovalYearPrimary indicationPARP-trapping potency
Olaparib (Lynparza)FDA2014BRCA-mutant ovarian; expanded to breast, prostate, pancreaticModerate
Rucaparib (Rubraca)FDA2016BRCA-mutant ovarian, prostateLow–moderate
Niraparib (Zejula)FDA2017Ovarian (all-comers; not BRCA-restricted)Low
Talazoparib (Talzenna)FDA2018BRCA-mutant breastHigh (most potent in class)
VeliparibNot approved—Failed pivotal trialsVery low
PamiparibNMPA (China)2021BRCA-mutant ovarianModerate

PARP-trapping potency hierarchy (descending): talazoparib > olaparib > niraparib > rucaparib > veliparib 1. Trapping — not just catalytic inhibition — is the dominant mechanism of synthetic lethality in BRCA-deficient cancers.

Mechanism

Catalytic inhibition

PARP1 binds single-strand breaks (SSBs) within milliseconds via its N-terminal zinc fingers, then auto-poly(ADP-ribosyl)ates (PARylates) itself and scaffold proteins (notably xrcc1), recruiting base-excision repair / SSBR machinery 2 3. PARP inhibitors occupy the NAD+-binding pocket (the ADP-ribosyl transferase domain) competitively, blocking PAR synthesis. Without PAR-mediated signalling, XRCC1 cannot be recruited to SSBs, and SSBR stalls.

PARP trapping (the dominant oncology mechanism)

More important than catalytic inhibition for synthetic lethality: when a PARP inhibitor binds PARP1 at an SSB, it stabilises the PARP1–DNA complex in a “trapped” state, preventing PARP1 dissociation from DNA 1. Trapped PARP1 complexes are far more cytotoxic than unrepaired SSBs alone — they block replication fork progression in S phase, converting SSBs to double-strand breaks (DSBs). HR-proficient cells can repair these DSBs via BRCA1/2-dependent pathways; HR-deficient (BRCA1/2-mutant) cells cannot — generating the synthetic-lethality window that is the basis of all approved oncology indications 4.

Mechanism componentCatalytic inhibitionPARP trapping
Blocks PAR synthesis?YesYes
Releases PARP1 from DNA?Prevents (trap)Prevents (trap)
Cytotoxicity in HR-deficient cellsModerateHigh (10–100× more than catalytic alone)
Cytotoxicity in HR-proficient cellsLowLow

PARP-trapping potency hierarchy

Talazoparib is 100-fold more potent as a PARP trapper than veliparib at equivalent inhibitor concentrations, despite similar catalytic IC50s 1. This explains why talazoparib showed superior efficacy in BRCA-mutant breast cancer (EMBRACA trial) relative to veliparib’s clinical failures, and is a critical structural-activity relationship for any future aging-context formulation work.

Aging-relevant biology

The NAD+–PARP1–sirtuin axis (the Cantó/Auwerx hypothesis)

PARP1 is the cell’s dominant NAD+ consumer under conditions of elevated DNA strand-break burden. Normal steady-state PARP1 activity is modest, but PARP1 hyperactivation — as can occur under chronic genotoxic stress — rapidly depletes cellular NAD+ 5. Mouchiroud 2013 demonstrated this causally in C. elegans: genetic or pharmacological inactivation of PARP1 (via AZD2281/olaparib or ABT-888/veliparib at 100 nM) increased NAD+ levels and extended worm lifespan by 15–23% in a sir-2.1 (sirtuin homolog)-dependent manner. Because sirtuin deacetylases (SIRT1, SIRT3, SIRT6) require NAD+ as a co-substrate (not just a cofactor), PARP1-mediated NAD+ depletion directly impairs sirtuin function:

  • SIRT1: reduced → hyperacetylation of PGC-1α, p53, NF-ÎşB → impaired mitochondrial biogenesis, increased inflammation
  • SIRT3: reduced → hyperacetylation of mitochondrial ETC subunits → reduced mitochondrial respiratory efficiency
  • Downstream: impaired mitochondrial function, increased ROS, self-amplifying DNA damage cycle

This axis has been demonstrated causally using PARP1 knockout mice and NMN/NAD+ supplementation models 6 — see also parp1 (verified) for the full mechanistic treatment. Note: Mouchiroud 2013 demonstrated the NAD+–PARP1–sirtuin axis using the PARP inhibitors AZD2281 (olaparib; 100 nM) and ABT-888 (veliparib; 100 nM), not PJ34 — see corrected footnote.

Aging hypothesis: in aged cells with elevated DNA damage burden and baseline PARP1 hyperactivation, PARP inhibitors might reduce constitutive NAD+ consumption, partially restoring the NAD+ pool available to sirtuins — a complementary strategy to NAD+ precursor supplementation. This is mechanistically coherent but has not been tested directly in a healthy-aging context. needs-human-replication no-mechanism (specific aging-context dose required to restore NAD+ without causing SSBR impairment)

DimensionStatus
Pathway (PARP1–NAD+–sirtuin axis) conserved in humans?yes
PARP1 hyperactivation documented in aged human cells?partial — elevated γ-H2AX and PAR in aged tissues; not directly quantified at the whole-body level
PARP inhibition restores NAD+/sirtuin in aged animals?not-tested (data from XPA/NER-deficient mice, not normal aging)
Replicated in humans for aging endpoint?no

Olaparib promotes senescence in murine macrophages (2025)

A 2025 GeroScience study (Kieronska-Rudek et al.) examined olaparib effects in senescent vs non-senescent murine macrophages (RAW264.7 cell line, replicative senescence model; passages 5–10 non-senescent vs 30–40 senescent) 7. At 1–30 µM for 72h (n=3–6 independent experiments per assay):

  • SA-β-gal (senescence marker): ~50% increase at 30 µM in non-senescent cells; 2.5-fold increase at 30 µM in senescent cells
  • p21: ~doubled at 30 µM in both cell types; baseline p21 was ~15% higher in senescent cells
  • Cell death: at 30 µM, senescent cells showed predominantly necrotic death (41% necrosis vs 39% apoptosis); non-senescent cells showed predominantly apoptotic death (56% apoptosis vs 32% necrosis by Annexin V/PI FACS)
  • Mitochondrial function: senescent cells had higher baseline bioenergetics; olaparib at 30 µM significantly suppressed mitochondrial function in both types

The authors note the concentrations tested (up to 30 µM) are at the high end of clinical relevance. Interpretation is bidirectional for aging: PARP inhibition at high doses preferentially kills senescent macrophages via necrosis (a potential senolytic-like effect), but also induces senescence markers in non-senescent immune cells (a pro-senescent harm). The study design — a single cell line model, not primary immune cells or in vivo — does not resolve which effect would dominate in vivo. contradictory-evidence needs-replication

Clonal hematopoiesis induction — a major aging-context safety signal

PARP inhibitors are now associated with an increased prevalence of clonal hematopoiesis (CH) in treated cancer patients, particularly through PPM1D, TP53, and other DNA-damage-response gene mutations 8. CH itself is an aging-associated state that confers elevated cardiovascular risk and increased risk of transformation to myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) 9. A case series of n=6 ovarian cancer patients on PARP inhibitor therapy (rucaparib, olaparib, or niraparib; median duration 18 months) demonstrated that clonal abnormalities and haematological abnormalities can be monitored and may partially reverse after PARPi cessation in some patients 9.

This is a critical translation barrier for aging use: in a cancer patient with expected survival of 2–5 years, short-course PARP inhibitor-associated CH induction may be an acceptable tradeoff. In a healthy 65-year-old seeking a 20+ year healthspan benefit, PARP inhibitor-induced CH acceleration could represent net harm by increasing 10–20 year MDS/AML risk. No analysis has quantified this tradeoff for the hypothetical healthspan use case. long-term-unknown

Clinical evidence

Oncology (established)

TrialAgentIndicationKey finding
SOLO-2 (N=295)OlaparibBRCA-mutant ovarian (maint.)mPFS 19.1 vs 5.5 mo, HR 0.30 (Pujade-Lauraine 2017, Lancet Oncol; not verified against primary source — cited via Lord 2017 review context) needs-replication
OlympiAD (N=302)OlaparibBRCA-mutant metastatic breastmPFS 7.0 vs 4.2 mo, HR 0.58, p<0.001 10
EMBRACA (N=431)TalazoparibBRCA-mutant breastmPFS 8.6 vs 5.6 mo, HR 0.54
PRIMA (N=733)NiraparibOvarian (all-comers)mPFS 13.8 vs 8.2 mo, HR 0.62

Individual compound pages will carry full trial details. Note: all approved indications require BRCA1/2 pathogenic variant (germline or somatic) as the synthetic-lethality pre-condition; the HR-deficient context is essential to the oncology mechanism.

Aging / healthspan (absent)

No completed RCT has tested a PARP inhibitor for an aging, longevity, healthspan, or senolytic primary endpoint in humans. The R25 recency search (2026-05-07) identified no recruiting or active-not-recruiting trials on ClinicalTrials.gov targeting these endpoints. All current active PARP inhibitor trials are oncology-focused. clinical-trials-active: 0 for aging indication.

Safety profile

Well-established in oncology (short-course):

  • Myelosuppression: anemia (25–35%), thrombocytopenia (15–25%), neutropenia (20–30%) — the most common Grade 3+ toxicities across the class
  • Nausea/vomiting: common (40–60%), typically manageable
  • Fatigue: very common
  • MDS/AML risk: ~1–2% cumulative at 2 years in oncology trials — elevated relative to background; likely higher with longer use
  • Clonal hematopoiesis induction: documented in ovarian cancer patients on PARPi therapy (Nuttall Musson 2024 9: n=6 cases, PPM1D and TP53 variants predominate; partially reversible after PARPi cessation); larger pharmacovigilance data (Marshall 2024 8 in prostate cancer) document increased CH prevalence in PARPi-treated patients

Unknown for aging-context (chronic, low-dose):

  • No chronic administration safety data in healthy older adults exists
  • The CH induction risk at sub-oncology doses is uncharacterized
  • Long-term MDS/AML risk in an otherwise healthy aging individual with 20+ year horizon is unquantified
  • Interaction with age-related renal decline (renally cleared agents) requires separate PK/PD characterization

#gap/long-term-unknown — no chronic aging-context safety data exists. The indication-split safety profile (well-established for oncology short courses; investigational and potentially harmful for aging chronic use) is reflected in the frontmatter safety-profile field.

Assessment: PARP inhibitors as a potential aging intervention

The mechanistic rationale is coherent: PARP1 hyperactivation is a documented consequence of the elevated DNA damage burden that accompanies aging, and the PARP1–NAD+–sirtuin–mitochondria axis is established in DNA repair-deficient models. However, the leap from “PARP1 hyperactivation contributes to aging” to “PARP inhibitors would extend healthspan” requires several untested assumptions:

  1. Dose calibration: oncology doses fully suppress PARP1 and cause myelosuppression; a sub-oncology dose sufficient to reduce constitutive age-related NAD+ drain without impairing acute SSBR has not been identified.
  2. HR status: PARP-trapping lethality requires HR deficiency; aged cells are not typically HR-deficient in a constitutive sense (though BRCA1 expression may decline with age in some tissues — unconfirmed).
  3. CH induction harm: the documented CH-inducing effect of PARP inhibitors represents a potentially serious antagonistic effect in a healthy aging context.
  4. NAD+ precursors as the simpler alternative: supplementing the NAD+ substrate pool (via NMN or NR — see nad-precursors) achieves the same downstream sirtuin-reactivation goal without the genotoxic burden of SSBR impairment or the CH-induction risk. The simpler strategy may outperform the more targeted one.

The PARP inhibitor aging hypothesis merits mechanistic investigation (low-dose in aged animals, without the NER-deficiency confound) but is not ready for human testing in a healthspan context.

ClassMechanismRelationship to PARPi
nad-precursorsNAD+ substrate supplementationSimpler route to same sirtuin-restoration goal; no CH-induction risk
senolyticsSelective clearance of senescent cellsPARPi may have weak senolytic activity (speculative; Kieronska-Rudek 2025)
base-excision-repairBER pathwayPARPi targets PARP1, a BER scaffold (verified pathway page)
homologous-recombinationHR pathwayPARPi efficacy requires HR deficiency (synthetic lethality mechanism)
sirtuinSIRT1/3/6 deacetylasesDownstream beneficiaries of NAD+ restoration

Limitations and open questions

  • #gap/needs-human-replication — all aging-axis evidence is from DNA repair-deficient animal models, not normal aging
  • #gap/no-mechanism — dose-response for NAD+-restoration vs SSBR-impairment in aged (non-NER-deficient) cells has not been established
  • #gap/long-term-unknown — no chronic low-dose PARP inhibitor safety data in healthy older adults; CH induction risk at sub-oncology doses uncharacterized
  • #gap/contradictory-evidence — olaparib promotes senescence in macrophages (Kieronska-Rudek 2025); unclear whether this is senolytic (kills pre-existing senescent cells) or pro-senescent (induces senescence in healthy cells)
  • #gap/needs-replication — single study (Kieronska-Rudek 2025) for the aging-macrophage senescence phenotype; n=3–6 independent experiments using RAW264.7 murine macrophage cell line only; primary immune cells and in vivo validation not yet done
  • #gap/dose-response-unclear — optimal dose for aging-hypothesis (if any) not identified; oncology doses cause myelosuppression

Cross-references

  • parp1 — verified protein page; canonical mechanistic anchor for NAD+-PARP1-sirtuin axis
  • base-excision-repair — verified pathway page; documents PARP1 scaffold role and Fang & Bohr 2016 NAD+-mitophagy axis
  • xrcc1 — verified protein page (R32a); PARP1’s primary scaffold partner at SSBs; recruited via PAR chains
  • brca1 — verified protein page; HR pathway anchor; the HR-deficiency requirement for synthetic lethality
  • homologous-recombination — pathway page for the synthetic lethality context
  • nad-precursors — sibling class; the alternative strategy for NAD+-sirtuin restoration
  • genomic-instability — primary targeted hallmark
  • deregulated-nutrient-sensing — secondary hypothetical target (via NAD+-sirtuin axis)
  • sirtuin — pathway page; downstream of PARP1–NAD+ axis

Footnotes

Footnotes

  1. doi:10.1126/scitranslmed.aaf9246 · review · Science Translational Medicine 2016 · Pommier Y et al. · mechanistic review of PARP-trapping vs catalytic-inhibition distinction; establishes talazoparib > olaparib > niraparib > rucaparib > veliparib trapping hierarchy; quantitative cellular protein-DNA complex assay data · local: not_oa (closed-access); cited from DOI lookup ↩ ↩2 ↩3

  2. doi:10.1038/nature03443 · bryant-2005-parp-inhibitor-brca2 · in-vitro + in-vivo · n=not-specified (tumour cell lines + xenograft) · model: BRCA2-deficient V-C8 hamster cells (V79 wild-type controls; V-C8+B2 BRCA2-rescued controls); 40 CD-1 nude mice for xenograft · PARP inhibitors NU1025 and AG14361 profoundly reduced survival of BRCA2-deficient cells at concentrations non-toxic to normal cells; V-C8 xenografts responded to 5-day AG14361 treatment (25–50 mg/ml) with tumour regression; V-C8+B2 xenografts showed no response; first demonstration of synthetic lethality between PARP inhibition and BRCA2 deficiency · specific fold-selectivity not stated as a single headline figure in this paper (Lord & Ashworth 2017 review cites “as much as 1,000-fold” for the class across both 2005 papers) · local: PDF verified 2026-05-07 ↩

  3. doi:10.1038/nature03445 · farmer-2005-parp-inhibitor-brca · in-vitro + in-vivo · model: BRCA1- or BRCA2-deficient embryonic stem (ES) cells (Cre-mediated knockout); xenograft in BALB/c-nude mice (40 mice; 2×10^6 ES cells injected) · PARP inhibitors KU0058684 (IC50=3.2 nM) and KU0058948 (IC50=3.4 nM) selectively killed BRCA1- or BRCA2-deficient ES cells; SF50 (dose causing 50% survival) = 35 nM for BRCA1-deficient; ~2 µM for wild-type — 57-fold sensitisation for BRCA1-deficient; 133-fold sensitisation for cells lacking both BRCA1 and BRCA2; xenograft: KU0058684 significantly blocked tumour formation from BRCA2-deficient ES cells (p=0.03 vs vehicle; p=0.01 vs wild-type treated); co-discovery of PARP–BRCA synthetic lethality with Bryant 2005 · local: PDF verified 2026-05-07 ↩

  4. doi:10.1126/science.aam7344 · review · Science 2017 · Lord CJ & Ashworth A · comprehensive review of synthetic-lethality mechanism, HR pathway, and PARP inhibitor clinical landscape as of 2017; BRCA-mutant cells were “as much as 1,000 times more sensitive” to PARPi than BRCA-wild type; talazoparib “approximately 100-fold more potent than niraparib” in PARP1 trapping; veliparib has “limited ability to trap PARP1”; SOLO-2 context included · local: PDF verified via PMC author manuscript 2026-05-07 ↩

  5. doi:10.1016/j.cell.2013.06.016 · mouchiroud-2013-nad-sirtuin-longevity · in-vivo · n=60–100 animals per condition (C. elegans, scored every other day) · model: C. elegans (N2 wild-type; pme-1(ok988) PARP-1 mutant; rrf-3(pk1426) RNAi-sensitised); primary mouse hepatocytes (Sirt1^L2/L2 conditional knockout); AML12 hepatocyte cell line · NAD+ supplementation via NR (500 µM) or NAM (200 µM) extended C. elegans lifespan +22% (NR; p=0.0004) and +16% (NAM; p=0.01), in a sir-2.1-dependent manner; PARP inhibition by AZD2281 (olaparib; 100 nM) extended lifespan +22.9% (p<0.001); ABT-888 (veliparib; 100 nM) extended lifespan +15% (p<0.05); both PARP inhibitor effects were pme-1-dependent and abrogated in sir-2.1(ok434) mutants; established NAD+–sirtuin axis and PARP1 as a major competing NAD+ consumer in vivo · Note: PARP inhibitors used were AZD2281 (olaparib) and ABT-888 (veliparib) at 100 nM — NOT PJ34 (an earlier tool compound not used in this paper) · local: PDF verified 2026-05-07 ↩

  6. doi:10.1016/j.cell.2014.03.026 · fang-2014-xpa-parp1-nad-mitophagy · in-vitro + in-vivo · model: XPA-deficient cells (NER-deficient; chronic PARP1 hyperactivation); XPA-deficient mice · PARP1 hyperactivation in XPA-deficient cells depleted NAD+; SIRT1 activity was reduced; mitophagy was defective; NAD+ repletion via NMN restored SIRT1, mitophagy, and mitochondrial function; PARP inhibitor (ABT-888/veliparib) at 5 µM partially rescued NAD+ and SIRT1 activity · Note: model is NER-deficient, not normal aging; extrapolation to normal aging requires caution · local: bronze OA; download pending ↩

  7. doi:10.1007/s11357-025-01679-6 · kieronskarudek-2025-olaparib-macrophage-senescence · in-vitro · n=3–6 independent biological experiments per assay (stated in figure legends) · model: murine macrophage cell line RAW264.7 (replicative senescence: passages 5–10 non-senescent vs passages 30–40 senescent); bone-marrow-derived macrophages were NOT used — only RAW264.7 · olaparib (0, 1, 3, 10, 30 µM for 72h) upregulated SA-β-gal (~50% increase at 30 µM in non-senescent; 2.5-fold increase in senescent) and p21 (doubled at 30 µM in both types); at 30 µM, senescent cells died predominantly by necrosis (41%) over apoptosis (39%), while non-senescent cells showed predominantly apoptotic death (56% vs 32% necrosis) by Annexin V/PI FACS; senescent cells showed higher baseline PARP1 expression (+50% full-length PARP1), higher PARylation (8.5× non-senescent), and higher baseline bioenergetics (ATP, mitochondrial respiration); authors note concentrations used (up to 30 µM) are potentially clinically relevant; authors suggest olaparib may exhibit senolytic-like activity in senescent immune cells but also exerts pro-senescent effects in non-senescent cells · Key limitation acknowledged by authors: all findings are from a murine cell line model; primary immune cells and in vivo validation needed · local: PDF verified 2026-05-07 needs-replication ↩

  8. doi:10.1002/pros.24712 · observational · n=not-specified (advanced prostate cancer patients) · model: humans (PARP inhibitor-treated prostate cancer) · PARP inhibitor treatment associated with increased prevalence and progression of clonal hematopoiesis; PPM1D mutations predominant; CH-associated MDS/AML risk concern documented · local: not_oa (closed-access); cited from DOI lookup no-fulltext-access ↩ ↩2

  9. doi:10.1038/s41375-023-02040-6 · case series · n=6 patients (stage III/IV high-grade ovarian cancer; mean age 67.3 years; median PARPi duration 18 months) · model: humans (PARP inhibitor-treated ovarian cancer patients: rucaparib n=1, olaparib n=3, niraparib n=2; referred to haematology for unexplained cytopenias/haematological abnormalities) · 5/6 patients had clonal pathology (CP) by m-NGS: PPM1D and TP53 variants most common; DNMT3A and SMC3A also detected; diagnoses ranged from CHIP to MDS-LB; 4/5 patients with CP showed resolution or improvement of haematological abnormalities after PARPi discontinuation, including 2 with MDS who became transfusion-independent; demonstrates partial reversibility of PARPi-associated CH/MDS in some patients · local: PDF verified 2026-05-07 ↩ ↩2 ↩3

  10. doi:10.1056/NEJMoa1706450 · robson-2017-olaparib-olympiad-breast · rct · n=302 (randomized 2:1: 205 olaparib, 97 standard therapy) · model: germline BRCA1/2-mutant HER2-negative metastatic breast cancer; median age 44 (olaparib) vs 45 (standard therapy) · olaparib 300 mg twice daily vs physician’s choice single-agent chemotherapy (capecitabine, eribulin, or vinorelbine); mPFS 7.0 vs 4.2 months (HR 0.58, 95% CI 0.43–0.80, p<0.001); ORR (blinded central review, measurable disease): 59.9% vs 28.8%; OS: HR 0.90 (95% CI 0.63–1.29; p=0.57) — no significant OS difference at data cutoff; primary analysis by stratified log-rank test, Kaplan-Meier · local: PDF verified 2026-05-07 ↩

Graph View

Backlinks