DNA Repair Inhibitors in Ovarian Cancer: Current Status and Future Strategies Stanley B. Kaye, MD Royal Marsden Hospital London, United Kingdom
DNA Repair Inhibitors in Ovarian Cancer: Current
Status and Future Strategies
Stanley B. Kaye, MDRoyal Marsden Hospital
London, United Kingdom
Why DNA Repair Inhibition?
• To selectively kill cancer cells with a defect in DNA repair
• In combination with DNA-damaging anticancer agents to overcome resistance
DNA Repair Inhibitors in Ovarian Cancer Agents with efficacy demonstrated in randomized trials:
• Olaparib: PARP inhibitor
• AZD 1775: WEE1 Kinase inhibitor
Agents in earlier clinical development targets include:• Afuresertib AKT
• Guadecitabine (SGI-110) DNMT inhibitor (epigenetic)
• PRIMA-1MET (APR-246) Mutant p53
• Others ATR/ATM; CHK 1/2, DNA-PK, etc
AND: Don’t forget targeted chemotherapy!
PARP Inhibition and Tumor-Selective Synthetic Lethality
PARP, poly(ADP)ribose polymerase;DSB, double-strand break; HR, homologous recombinationSSB, single-strand break;NHEJ, non-homologous end joining
PARP
DNA damage (SSBs)
DNA replication(accumulation of DNA DSBs)
Normal cellwith functional HR pathway
HR-deficient tumor cell (eg, BRCA1/2-/-)
Cell survival Cell death
HR-mediated DNA repair
Impaired HR-mediated DNA repair (NHEJ etc.)
Tumor-selective cytotoxicity
PARP inhibition
PARP inhibitors can also trap cytotoxic PARP-DNA complexes; clinical relevance unclear.
Murai J, et al. Cancer Res. 2012;72(21):5588-5599.
BRCA1 Germline
8%
BRCA2 Germline
6%
BRCA1 Somatic
3%BRCA2
Somatic3%
BRCA1 Methylation
11%
EMSY Amplification
6%PTEN Loss
5%Other HRD7%
CCNE1 Amplification
15%
MMR Germline
2%
Other 34%
The Incidence of BRCA Mutations in High-Grade Serous Ovarian Cancer
• BRCA1/2 germline mutation 14%
• BRCA1/2 somatic mutation 6%
• Total 20%
The Cancer Genome Research Network. Nature. 2011;474(7353):609-615.
BRCA germline mutation testing should be routine…? somatic testing too
Not HRD HRD
Approx 50% of HGSOC could be candidates for PARPi
Phase I trial of KU59436 (olaparib) indicated excellent tolerance and expansion in 50 BRCA
patients showed 46% response
Fong P, et al. N Engl J Med. 2009;361(2):123-134.Fong P, et al. J Clin Oncol. 2010;28(15):2512-2519.
Preclinical
Farmer H, et al. Nature. 2005;434(7035):917-921.
Exquisite preclinical efficacy of PARPi in BRCA deficient ES cells
“this is nothing like chemotherapy”
Bryant HE, et al. Nature. 2005;434(7035):922-926.
Olaparib, Chapter 1, 2005-9Early clinical trials(Phase I incl. IB)
KU-0058948IC50 = 3.4 nM
1250 fold difference in SF50 between BRCA2 -/- and +/+
Clonogenic survival curves with inc. concentration of KU 58948
Olaparib 400 mg PO bidRandomized 1:1
PlaceboPO bid
Patients:• Platinum-sensitive high-grade serous
ovarian cancer • 2 previous platinum regimens • Last chemotherapy was platinum-
based to which they had a maintained PR or CR prior to enrolment
• Stable CA-125
Treatment until
disease Progression
Study aim and design
Total of 265 recruited: Of which 51% had germline/somatic BRCA mutation
Primary endpoint : PFS
Ledermann J, et al. N Engl J Med. 2012;366(15):1382-1392.
Olaparib, Chapter 2, 2010-14Randomized Trial of Maintenance Olaparib in Platinum-
Sensitive Relapsed Ovarian Cancer (Study 19)
HR = 0.35 (95% CI: 0.25, 0.49)
P<.00001
Number at Risk
Olaparib 136 106 53 24 7 0
Placebo 129 72 24 7 1 0
Study 19: Met PFS Primary Endpoint
0
0.6
0.8
0.9
0
0.1
0.2
0.3
0.4
0.5
0.7
1.0
3 6 9 12 15
Probability of
PFS
Time From Randomization, Months
Placebo
Olaparib
OlaparibN = 136
PlaceboN = 129
Median (95%CI)
8.4 months(7.4, 11.5)
4.8 months(4.0, 5.5)
Ledermann J, et al. N Engl J Med. 2012;366(15):1382-1392.
0Time From Randomization, Months
0
1.0
Prop
ortio
n of
Pat
ient
s Pr
ogre
ssio
n Fr
ee
3 6 9 12 15
0.90.80.70.60.50.40.30.20.1
Olaparib BRCAmPlacebo BRCAm
Number at RiskOlaparib BRCAmPlacebo BRCAm
74 59 34 15 5 062 35 13 2 0 0
BRCAm (n = 136)Olaparib Placebo
Events: total patients (%) 26:74 (35.1) 46:62 (74.2)Median PFS, months 11.2 4.3
HR = 0.1895% CI (0.10, 0.31); P<.00001
BRCAwt, BRCA wildtype*Includes patients with germline and/or somatic mutations; †patients were treated until disease progression
Randomized treatment†
Ledermann J, et al. Lancet Oncol. 2014;15(8):852-861.
PFS in Patients With a BRCA Mutation*
Overall Survival in Patients With BRCA Mutation
• 14/62 (22.6%) placebo patients switched to a PARP inhibitor
0Time From Randomization, Months
048
1.0
Pro
po
rtio
n o
f P
atie
nts
Aliv
e
3 6 9 12 15 18 21 24 27 30 33 36 39 42 45
Number at risk
62 62 58 52 50 46 39 36 33 29 29 27 21 12 4Placebo BRCAm
74 71 69 67 65 62 57 54 50 48 39 36 26 12 7Olaparib BRCAm
Randomized treatment
Placebo BRCAm
Olaparib BCRAm
0.90.80.70.60.50.40.30.20.1
BRCA mut (n = 136)Olaparib Placebo
Deaths: total patients (%) 37:74 (50.0) 34:62 (54.8)Median OS, months 34.9 31.9
HR = 0.7395% CI (0.46, 1.19)
P = 0.192
Note: Only 58% of maturity – next analysis – this year ? impact of long-term survivors
Ledermann J, et al. Lancet Oncol. 2014;15(8):852-861.
Randomized Trial of Olaparib as Maintenance Therapy in Platinum-Sensitive Sporadic Ovarian Cancer
Trial positive for primary endpoint (PFS). But overall survival impact less clear.Does this reflect cross-over (23%), or too early analysis? When patients progress on olaparib do they still respond to platinum?
What do we really know about PARPi (and platinum) resistance?
In separate retrospective analysis of chemo given post olaparib:• 19/48 (40%) response to platinum • Med PFS. 22 weeks• No evidence of secondary BRCA mutations in 6 cases.
Ang JE, et al. Clin Cancer Res. 2013;19(19):5485-5493.
Resistance to PARP Inhibitors
Fojo T, et al. Cancer Discov. 2013;3(1):20-23.Ceccaldi R, et al. Cancer Res. 2015;75(4):628-634.Jaspers JE, et al. Cancer Discov. 2013;3(1):68-81.
– Secondary BRCA mutation– P-glycoprotein-based
enhanced drug efflux– Reduced 53BP1, partially
restoring HR– NER pathway alterations
– And why do a minority of cases (up to 20%) stay in remission long-term?
– Is this all due to tumorheterogeneity?
• Is likely to be multifactorial; factors to consider include:
Answers will require tumor samples from patients progressing
on PARPi
Meanwhile…..Olaparib in Advanced Recurrent BRCAm Ovarian Cancer
• From the Kaufman et al paper, data on subgroup of 137 patients who received ≥3 lines of chemo presented to FDA for accelerated approval
• Response rate 34%; response duration 7.9 months
Total of 300 patients treated in 6 trials including:
• Initial phase I/II trials - Fong et al, NEJM 2009, JCO
2010, Audeh et al Lancet 2010• Randomized trial vs PLD
- Kaye et al, JCO 2012• Bioavailability and scheduling
studies - Capsule » tablet, cont. v
intermittent, Mateo et al, EJC
2013• Non-randomized, multiple
BRCAm disease - Kaufman et al JCO 2015
Fong PC, et al. N Engl J Med. 2009;361(2):123-134. Fong PC, et al. J Clin Oncol. 2010;28(15):2512-2519. Audeh MW, et al. Lancet. 2010;376(9737):245-251. Kaye SB, et al. J Clin Oncol. 2012;30(4):372-379. Mateo J, et al. Eur J Cancer.
2013;49(Suppl2): Abstract 801. Kaufmann B, et al. J Clin Oncol. 2015;33(3):244-250.
Status of Olaparib – January 2016
Europe – approved as maintenance treatment for platinum sensitive relapsed BRCA m ovarian cancer – patients in remission following platinum-based therapy
USA – approved as monotherapy for patients who have received ≥3 lines of chemotherapy• Not approved as maintenance therapy• Approval also for companion diagnostic (Myriad Genetics
BRCA analysis CDx)
a) As capsules (400 mg bd)
b) As tablets (300 mg bd)• Confirmatory studies ongoing
• Define activity in sporadic ovarian cancer and other cancers, eg, prostate
• Assess PARP inhibitors other than olaparib (rucaparib, niraparib, BMN-673)
• Develop robust predictive biomarker• Test novel combinations (with P13K or
angiogenesis inhibitors, etc)• Monitor long-term toxicity• Understand mechanisms of PARPi resistance
PARP Inhibitors – What Are The Next Steps?
Single-Agent Activity for PARP Inhibitors in Ovarian Cancer
Drug BRCA Mutation Positive BRCA Wildtype and Unknown
n% RespRECIST
Resp. Duration n
% RespRECIST
RespDuration Reference
Olaparib >100 (most plat resist)
30 to 60%
7-10 months 46 24%(50% in plat
sens4% in plat
resist)
7m Fong et al. JCO 2010
Kaye et al. JCO 2012Gelmon et al. Lancet
Oncology 2011
Rucaparib 39 (all plat sens)
69% >9 months 74(HRD high)
62(HRD low)
30%
13%
7m
4m
McNeish et alASCO 2015
Niraparib 20 (9 plat sens)
45% 11 months 20 15% 5m Sandhu et al. Lancet Oncology 2013
BMN 673 28 (22 plat sens)
68% >6 months Ramanthan et al EJC 2013 suppl 3 LBA 29
Fong PC, et al. J Clin Oncol. 2010;28(15):2512-2519. Kaye SB, et al. J Clin Oncol. 2012;30(4):372-379. Gelmon KA, et al. Lancet Oncol. 2011;12(9):852-861. McNeish IA, et al. J Clin Oncol. 2015;33(suppl): Abstract 5508. Sandhu SK, et al. Lancet Oncol. 2013;14(9):882-892. Ramanathan R, et al. Eur J Cancer. 2013;49(Suppl2): Abstract 29.
McNeish IA, et al. J Clin Oncol. 2015;33(suppl): Abstract 5508.
• HRD causes genome wide loss of heterozygosity (LOH), which can be measured by genome profiling using NGS
• Algorithm developed for LOH score (high/low), ie, BRCA-like signature, with LOH cut off derived from OS data on cohort of 309 platinum-treated patients
Progression-free survival by HRD molecular subgroup
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1.00.90.80.70.60.50.40.30.20.1
0
Time, Months
BRCAmu
t
BRCA-like
Chromosome No.
Biomarker Negative
BRCAwt
BRCAmut
BRCA-likeBiomarker Negative
• Correlation with efficacy of rucaparib in Phase II trial –ARIEL 2
BRCA-like : HRD high PFS: 7mBiomarker neg: HRD low PFS 4m
Homologous Recombination Deficiency (HRD) AssayDo We Have One?
Haluska P et al, NCI/EORTC/AACR 2014 (Eur J Cancer. 2014;50(Suppl6): Abstract 214.)
Developed HRD score incorporating 3 components:
HRD score is sum of LOH + TAI + LST scores- Presented evidence of correlation between HRD score and in
vitro/in vivo response to niraparib in 106 tumor samples – Clinical data in ovarian cancer awaited
• Loss of heterozygosity (LOH)• Telomeric allelic imbalance (TAI)• Large-scale state transitions (LST)
Thus:- Two assays under further evaluation, as key elements
in 2 ongoing randomized maintenance trials, with niraparib and rucaparib in sporadic and BRCAm-
associated ovarian cancer
Homologous Recombination Deficiency (HRD) AssayDo We Have Another?
PARP Inhibitor – Combination StrategiesAim: Enhance activity of PARPi by increasing HRD in treated cellsPre-clinical and clinical data with:
• Antiangiogenic agents1
• P13K/AKT pathway inhibitors2
Pre-clinical data with:• Wee1 Kinase inhibitors3
• ATR inhibitors4
1. Chan N, et al. Clin Can Res. 2010;16(18):4553-4560. 2. Rehman FL, et al. Cancer Discov. 2012;2(11):982-984. 3. Karnak D, et al. Clin Canc Res. 2014;20(19):5085-5096. 4. Huntoon C, et al. Canc Res. 2013;73(12):3683-3691.
Antiangiogenic Agents/PARP Inhibitors• Complementary targets/mechanisms
of action
• Potential enhancement of sensitivity to PARPi by increasing HRD through changes in oxygenation caused by antiangiogenic agent
• Bevacizumab/olaparib – phase I trials confirmed feasibility and randomised trial planned
• Cediranib/olaparib – positive randomised trial presented at ASCO 2014 – further randomised trials (incl. maintenance) ongoing
Dean E, et al. Br J Cancer. 2012;106(3):468-474.Liu JF, et al. Lancet Oncol. 2014;15(11):1207-1214.
Will benefit mainly be in patients with BRCA wildtype?
PARP Inhibitor Plus PI3K Inhibitor
• Phase I trials now underway, including olaparib plus AZD5363
– Initial data encouraging with no overlapping toxicity
Juvekar A, et al. Cancer Discov. 2012;2(11):1048-1063.Rehman FL, et al. Cancer Discov. 2012;2(11):982-984.
• Preclinical data in TNBC cells demonstrate that P13K inhibition suppresses BRCA1/2expression and enhances sensitivity to PARP inhibition, partly through activation of ERK and transcription factor ETS1
Ibrahim YH, et al. Cancer Discov. 2012;2(11):1036-1047.
Control –BKM 120 –Olaparib –BKM plus olaparib –
DNA Repair Inhibitors in Ovarian CancerClinical data so far available:a) As monotherapy to selectively kill cancer cells with defects in DNA
damage response or DNA repair.
b) In combination with DNA-damaging cytotoxic, eg, carboplatin, to increase its efficacy by inhibiting DNA-repair.
PARPi WEE 1
*
* Oza et al. Lancet Oncol. 2015;16(1):87-97. No clear benefit for concurrent olaparib
WEE-1 Kinase as a Target in Ovarian CancerCombination with chemotherapy
• TP53 mutated in >90% of HGSOC, resulting in loss of regulation of G1/S cell cycle point.
• TP53 mutant tumours more dependent on G2/M checkpoint to repair damaged DNA – this is regulated by Wee-1 kinase.
M
S
G1G2
CDK-cyclin
p53
Wee1
AZD1775
• AZD1775 is a small molecule Wee-1 kinase inhibitor, which acts synergistically with chemo, including platinum, particularly in TP53-mutant cell lines and xenografts through deregulation of G2/M checkpoint.
• Clinical trial data available in both platinum-sensitive and resistant disease
Slide courtesy of Amit Oza
Randomized Trial of Carbo/Paclitaxel ± AZD1775 in Platinum Sensitive Relapsed Ovarian Cancer
RecurrentHGSOC
Loss of function p53
mutation
≥6 months platinum-free interval
n = 59Carbo AUC 5Paclitaxel 175 mg+ AZD 1775 225 mg bd2.5 days only on each 21-day cycle
Carbo AUC5Paclitaxel 175 mg + placebo bd for 2.5 days only on each 21-day cyclen = 62
358 tumors sequenced using Roche p53 Amplichip
192 (53.6%) eligible mutations113 (31.6%) wildtype53 (14.8%) ineligible mutation
Primary endpoint PFS (using enhanced RECIST v 1.1) – volumetric, Secondary endpoint PFS (using RECIST v 1.1) – longest diameters
Oza AM, et al. J Clin Oncol. 2015;33(Suppl): Abstract 5506.
Progression-Free Survival by Enhanced RECIST RECIST 1.1
AZD1775 + P/C n = 59 P/C + Placebo n = 62
PFS events, n (%)
35 (59.3) 37 (59.7)
Median PFS, weeks
34.14 31.86
HR = 0.6380% CI 0.45, 0.8995% CI 0.38, 1.06
P = .080
Time (months)
AZD1775 + P/C n = 59 P/C + Placebo n = 62
PFS events, n (%)
34 (57.6) 32 (51.6)
Median PFS, weeks
42.86 34.86
HR = 0.5580% CI 0.39, 0.7995% CI 0.32, 0.95
P = .030Significant improvement in PFS
Oza AM, et al. J Clin Oncol. 2015;33(Suppl): Abstract 5506.
Response Rates by Enhanced RECIST, RECIST 1.1 and CA-125 Levels
AZD1775 + P/Cn = 59
P/C + Placebon = 62
Enhanced RECIST/CA-125 responseORR, % (95% CI) 81.4 (69.1, 90.3) 75.8 (63.3, 85.8)
RECIST 1.1 only responseORR, % (95% CI) 66.1 (52.6, 77.9) 51.6 (38.6, 64.5)
CA-125 only responseResponse rate, % 81.4 74.2
Trend to higher response rate using all criteria
Oza AM, et al. J Clin Oncol. 2015;33(Suppl): Abstract 5506.
Safety and Tolerability
AE categoryAZD1775 + P/C
n = 59P/C + Placebo
n = 60
AEs* All Grades Grade ≥3 All Grades Grade ≥3
Diarrhoea, n (%) 44 (74.6) 6 (10.2) 22 (36.7) 2 (3.3)
Vomiting, n (%) 37 (62.7) 6 (10.2) 16 (26.7) 1 (1.7)
Anemia, n (%) 31 (52.5) 12 (20.3) 19 (31.7) 4 (6.7)
Neutropenia, n (%) 26 (44.1) 21 (35.6) 24 (40.0) 20 (33.3)
Thrombocytopenia, n (%) 21 (35.6) 12 (20.3) 16 (26.7) 7 (11.7)
AE, adverse event*AEs tabulated with frequency >35% in the AZD1775 + P/C arm
No major differences in treatment tolerability
Oza AM, et al. J Clin Oncol. 2015;33(Suppl): Abstract 5506.
WEE1 Inhibitor/Chemotherapy in Platinum-Resistant Relapsed Ovarian Cancer • Phase II trial of AZD1775 plus carboplatin
AUC5 in 22 eval. patients with p53-mutated cancer, platinum-refractory or resistant (<3 m) to first-line carbo/paclitaxel– 6/22 (27%) had RECIST PR with med PFS of 10.9 m– 9/22 (41%) had RECIST SD with med PFS of 5.3 m
• Further phase II studies of AZD 1775/chemo underway in platinum-resistant disease
Leijen S, et al. J Clin Oncol. 2015;33(Suppl): Abstract 2507.
WEE-1 Kinase as a Target in Ovarian Cancer – Other Approaches
Monotherapy • Wee-1 kinase inhibition abrogates G2-M arrest, propelling cells
into premature mitosis and cell death.• In vitro studies and PD data from single agent Phase I trial,
suggest AZD1775 can induce cytotoxicity based on DNA damage.
• In phase I monotherapy trial, 2 responses seen (in patients with BRCA mutation) – further study planned in PARPi-failures
Combination with olaparib• Wee-1 kinase inhibition also induces HRD• Wee-1 kinase inhibition sensitises BRCA proficient cells to
PARP inhibition• Phase I trial of olaparib/AZD1775 underway
Do K, et al. J Clin Oncol. 2015;33(30):3409-3415.
Strategies to Circumvent Platinum Resistance
Mechanisms underlying platinum resistance may include:
Enhanced repair of DNA
damageFailure to recognize
DNA damage
Failure to trigger apoptosis
WEE1 inhibitorAKT inhibitor
Demethylatingagent
Target mutant p53, eg, APR-246
Therapeutic approaches may include
Targeting AKT as Strategy for Resistance Reversal
P13 kinase/AKT pathway activation (mutations /amplification) seen in up to 40% of ovarian cancer – in patients with ascites, correlation with drug resistance is noted.Afuresertib (GSK 2110183)- Pan AKT inhibitor, preclinically
reverses platinum resistance in
vitro
- Phase I study combined with paclitaxel/carboplatin included 23 patients with platinum resistant/refractory disease
Predictive biomarker: p- S6K/S6K ratio is a downstream readout of activation of the PI3 kinase pathway
Activation of PI3 kinase in tumour cells isolated from ascites, and response to subsequent chemotherapy
Carden C. et al. Mol Can Ther. 2012;11(7):1609-1617.
Blagden et al. Eur J Cancer. 2014;50:56.
For patients treated at MTD (125 mg)RECIST resp 50%GCIG resp 50%
Demethylation as a Strategy for Resistance Reversal
• Platinum resistance due to failure to recognise DNA damage through methylation of HMLH1
Brown R, et al. Oncogene.
1997;15(1):45-52.
• Demethylating agent, decitabine enhances platinum response in vitro/vivo
• Clinical correlation established between methylation acquisition and drug resistance
Plumb JA, et al. Cancer Res.
2000;60(21):6039-6044.
Gifford G, et al. Clin Can Res.
2004;10(13):4420-4426.
BUT: in randomised clinical trial, decitabine does not enhance (actually reduces) efficacy of carboplatin and trial stopped.
Glasspool RM, et al. Br J Cancer.
2014;110(8):1923-2014.
Will guadecitabine (SG110) do better?- Decitabine analogue, superior in vitro
- Ongoing randomized trial with carboplatin in platinum resistant disease (Fleming et al. AACR. 2014.)
LAB
CLINIC
Targeting p53 as Resistance Reversal Strategy• Cells with mutant p53 less able to trigger apoptosis
following chemo, eg, platinum
• APR-246 (PRIMA-1) small molecule pro-drug restoring mutant p53 to wildtype configuration, sensitising ovarian cancer cells to platinum
Mohell N, et al. Cell Death Dis. 2015;6:e1794. Gourley C, et al. J Clin Oncol. 2015;33(Suppl): Abstract TPS5605.
Ongoing clinical trial combining APR-246 with PLD/carboplatin (PISARRO) shows encouraging safety and activity profile, with RECIST response in all (8) patients treated so far (platinum-sensitive disease)
And Finally, Don’t Forget: Antibody-Targeted Chemotherapy in Ovarian Cancer!
Target “Warhead” Antibody Drug Conjugate CompanyFolate receptor1
MaytansinoidDM4
IMGN853 Immunogen
NaPi2B2 MMAE DNIB 0600 GenentechMesothelin3 MMAE DMOT 4039A GenentechTissue factor4 MMAE HuMax-TF GenmabsMUC 165 MMAE DMUC 5754 GenentechPTK76 MMAE PF 0664702 Pfizer
• Examples of efficacy in platinum-resistant ovarian cancer in all cases• Randomized studies planned/ongoing
MMAE, monomethyl auristatin
1. Moore K, et al. J Clin Oncol. 2015;33(Suppl): Abstract 5518. 2. Gordon MG, et al. J Clin Oncol. 2013;31(Suppl): Abstract 2507. 3. Weekes CD, et al. J Clin Oncol. 2014;32(5S): Abstract 2529. 4. Lassen UN, et al. J Clin Oncol. 2015;33(Suppl): Abstract 2570. 5. Liu J, et al. Cancer Res. 2013;73(24 Suppl): Abstract LB-290. 6. Tocher et al, ESMO/ECC. 2015.
Summary – DNA Repair Inhibitors
• In ovarian cancer, targeting DNA repair pathways, using both single agent and combination approaches, represents an important step forward in therapy
• Understanding clinical resistance to platinum (using paired samples) is a key priority for further research
Acknowledgements • Johann de Bono• Tim Yap• Joo Ern Ang• Peter Fong• Craig Carden• Martin Gore• Susanna Banerjee• Amit Oza, Toronto
• All the research nurses, clinical fellows and data managers in the DDU
• Support from CRUK, ICR and Biomedical Research Centre at RMH
• Clinical collaborators in Europe, USA, Australia• Colleagues at AZ, Clovis, and Tesaro