1. Homologous recombination deficiency (HRD) of high grade serous ovarian tumors from the NOVA Phase III clinical study
Keith Wilcoxen,1 Christopher Neff,2 Victor Abkevich,2 Joshua Timothy Jones,2 Kathryn Kolquist,2 Michael Mirza,2 Jerry Lanchbury,2 Keith Mikule,1 Shefali Agarwal,1 Anne-Renee Hartman,2 Alexander Gutin,2 and Kirsten Timms.2
1TESARO, Inc., Waltham, MA, USA 2Myriad Genetics, Inc., Salt Lake City, UT, USA
Introduction
Evaluation of Phase 3 (NOVA) patient tumors
non-gBRCAmut (n=211 analyzed to date)
DDR
Ovarian cancer is characterized by a high degree of genomic instability caused by deficiencies in DNA repair.
Cells can develop these homologous recombination (HR) deficiencies through the loss of function or inactivation of genes involved in DNA repair, such as BRCA1 and BRCA2.
Ovarian cancers known to have these HR deficiencies have been shown to benefit from therapy with DNA-damaging agents, such as platinum and PARP-inhibitors.
Homologous recombination deficiency (HRD = defined as the sum of LOH, LST and TAI genomic defects), assessed across >500 ovarian tumors exhibits a bio-modal distribution allowing a clear differentiation of HR proficient and HR deficient tumors.
Genome wide analysis was conducted on tumors obtained from patients enrolled in the NOVA study, a phase 3 clinical trial evaluating the PARP inhibitor niraparib as a maintenance treatment in patients with platinum sensitive ovarian cancer.
Homologous recombination deficiency (HRD), sequence analysis of 43 genes involved in DNA damage response and other measures of genomic instability were evaluated.
LOH score only
Niraparib in 2nd Line (Recurrent) Ovarian Cancer Maintenance (NOVA)
BRCA mutant
BRCA wt
LOH only cutoff of ≥7
High Grade Serous Ovarian Cancer, Platinum Sensitive, Relapsed
Number of tumors
Response to Platinum Treatment n=490
Non-gBRCAmut
gBRCAmut
2:1 Randomization
2:1 Randomization
Niraparib 300 mg QD
Placebo
Niraparib 300 mg QD
Placebo
0-1
2-3
4-5
6-7
8-9
10-11
12-13
14-15
16-17
18-19
20-21
22-23
24-25
26-27
28-29
30-31
32-33
34-35
36-37
38-39
40-41
42-43
44-45
46-47
48-49
50-51
52-53
54-55
56-57
58-89
60-61
62-63
64-65
66-67
68-69
70-71
72-73
74-75
76-77
78-79
80-81
82-83
84-85
86-87
88-89
90-91
92-93
94-95
96-97
98-99
100
LOH Score
n=207
n=103
n=120
n=60
BRCA deficient
BRCA intact
Endpoint Assessment
Endpoint Assessment
HRD score = LOH+LST+TAI
HRD cutoff of ≥ 42
Primary Endpoint:
PFS; >90% power to detect 4.5 month improvement (HR 0.50 )
Non-gBRCAmut cohort endpoint assessed hierarchically to control type 1 error: HRD + population first, followed by entire population
#Number of tumors
BRCA mutant
BRCA wt
Number of tumors
gBRCAmut (n=136 analyzed to date)
HRD score
Homologous recombination proficient (HRD-)
Homologous recombination deficient (HRD+)
LOH score only
HRD Score
≥7 = cutoff
95% of BRCAmut
HRD distribution in NOVA tumors (n=347 analyzed)
Number of tumors
HRD score = LOH+LST+TAI
gBRCAmut
tBRCAmut
HRD+ BRCAwt
HRD-
Conclusions
Methods
An assay has been developed that provides a quantitative continuous measure of genomic scarring and BRCA1/2 sequencing from tumor tissue in one test.
Genomic analysis of 347/490 patient tumor samples in the NOVA study has been conducted, indicating 100% of the gBRCAmut cohort and 55% of the non-gBRCAmut cohort are HRD positive as defined by the Myriad HRD test.
The use of three algorithms of DNA damage allow a clearer differentiation of the HRD population in ovarian cancer than the use of the single LOH algorithm.
Deleterious or suspected deleterious mutations, with confirmed loss of both tumor alleles, were detected in 12 additional DNA damage response genes in 35 tumors.
HRD scores in tumors with DDR gene mutations showed a similar range of HRD scores to that observed in BRCA1/2 mutants (25-82 compared to 25-88).
Patients: The NOVA study is a Phase 3, multicenter, randomized, double-blind, placebo-controlled study of niraparib as maintenance in platinum sensitive ovarian cancer patients who have either gBRCAmut or a tumor with high-grade serous histology and who have responded to their most recent chemotherapy containing a platinum agent.
Sample Analysis: DNA was extracted from formalin fixed paraffin-embedded (FFPE) tumor tissue and used to create libraries that were hybridized to a custom Agilent SureSelect capture array carrying probes for 54,091 single nucleotide polymorphism sites distributed across the human genome, as well as probes targeting 43 genes involved in DNA repair, including BRCA1 and BRCA2. The captured and enriched DNA was sequenced on an Illumina HiSeq 2500 sequencer. Sequences covering SNP positions were used to generate allelic imbalance profiles. Measures of genomic instability, including determination of a HRD score (integer value of 0-100), were calculated using allelic imbalance profiles and determination of loss of heterozygosity by ASCN.
HRD Score: The HRD score is the unweighted sum of LOH (number of subchromosomal LOH regions longer than 15 Mb), TAI (number of regions with allelic imbalance that extend to one of the subtelomeres but do not cross the centromere), and LST (the number of break points between regions longer than 10 Mb after filtering out regions shorter than 3 Mb).
Tumor samples were defined as HR deficient if they produced a score ≥42 and/or BRCA1/2 mutation.
LOH Score
# of tumors
HRD score = LOH+LST+TAI
≥ 42 = cutoff
95% of BRCAmut
HRD Score
Number of tumors
Shift to higher
scores
non-gBRCAmut cohort
gBRCAmut cohort
Total
211
136
tumor BRCA1/2mut
18% (39/211)
tumor BRCA1mut
71% (97/136)
HRD+ (BRCAwt)
37% (77/211)
tumor BRCA2mut
29% (39/136)
HRD-
45% (95/211) NA
HRD+
55% (116/211)
100% (136/136)
HRD Score (LOH+TAI=LST)
Analysis of LOH score only from tumors obtained from gBRCAmut cohort patients indicates a broad score distribution.
Utilizing the sum of three algorithms to score HRD (LOH, LST, TAI) results in a shift to higher HRD scores for BRCAmut tumors in NOVA.
A cutoff of 7 or greater for LOH score alone captures 95% of gBRCAmut tumors in NOVA.
A cutoff of 42 or greater for HRD scores captures 95% of gBRCAmut tumors in NOVA.
HRD+ tumors are defined as those with an HRD score of ≥ 42 or a deleterious or suspected deleterious mutation in BRCA1/2 as detected from tumor tissue
Patient eligibility criteria for the NOVA study are consistent with enrichment for patients with tumors having high HRD scores due to platinum sensitivity
Patients with HRD positive tumors consist of 100% of the patients in the gBRCAmut cohort and 55% of the patients in the non-gBRCAmut cohort based on this preliminary analysis of 347 pts.
References