Page 137 - Read Online

Murthy et al. Cancer Drug Resist 2019;2:665-79 I http://dx.doi.org/10.20517/cdr.2019.002 Page 671

PARP Veliparib Rucaparib Olaparib Niraparib Talazoparib
Targets PARP1 PARP1 PARP1 PARP1 PARP1
PARP2 PARP2 PARP2 PARP2 PARP2
PARP3 PARP3

Figure 2. PARP inhibitor targets

PARP trapping potency varies considerably among the PARPis, with talazoparib demonstrating the highest
PARP trapping potency [50,51] . Olaparib may be a weaker PARP trapper than talazoparib, and veliparib may
be a weaker PARP trapper than olaparib, based mostly on in vitro studies [48,50] . However, it is important
to note that efficacy and monotherapy activity of different PARPis does not correlate clearly with PARP
trapping potency. Nevertheless, an individual PARPi’s trapping potency may correlate with the maximum
tolerated dose and the tolerability of the drug in combination therapy (both are inversely correlated with
PARP trapping potency) [50,51] .

The PARP family of enzymes consists of at least 17 members, of which PARP1 and PARP2 have been clearly
found to participate in DNA repair. PARP1 is the best characterized and most abundant. More recently,
[52]
PARP3 was found to be involved in the repair of single-strand DNA breaks , among other functions.
Detailed analysis of the differences between known PARP family members is beyond the scope of this
review, and is an emerging area of research. PARP inhibitor targets include PARP1, PARP2, and PARP3; all
of the clinical PARPis target PARP1 and PARP2, with some additionally targeting PARP3 [Figure 2].

It is also important to note that PARP1 and PARP2 have other functions beyond involvement in DNA
break repair, which include roles in transcription, replication, modulating chromatin structure, and
stabilization of replication forks. Hence, PARP inhibition has complex repercussions on cellular stability,
much of which remains to be elucidated.

Clinical findings (activity, toxicity, pharmacological features)
It is difficult to directly compare the activity of different PARPis since head-to-head studies are lacking.
However, similarly designed clinical trials evaluating different PARPis have tended to show similar
results. For example, the phase 3 ARIEL3 and ENGOT-OV16/NOVA trials evaluating rucaparib and
niraparib, respectively, as maintenance treatment in platinum-sensitive recurrent ovarian cancer have
demonstrated comparably improved PFS in the PARP inhibitor arms compared to placebo. The OlympiAD
and EMBRACA trials in metastatic breast cancer, which evaluated olaparib and talazoparib, respectively,
also showed a similarly improved PFS in the PARP inhibitor arms compared to physician’s choice
chemotherapy.

Differences in toxicities between the PARPis, however, have emerged from these as well as other clinical
trials. One cannot exclude that differences reflect not only the dosing of the agent but patient selection and
prior treatment exposure to genotoxic agents. Proteome-wide profiling of the clinical PARPis also suggest
[53]
that specific PARPis may have differing off target effects , but it is not yet known whether these differences
translate to unique toxicities. Common toxicities for all PARPis are fatigue, gastrointestinal toxicities
(nausea/vomiting, abdominal pain, diarrhea) and cytopenias. Most of these are mild (grade 1-2). Overall,
grade 3 or greater toxicities occurred in approximately 35%-56% of patients treated with the approved
PARPis, of which a majority were hematological toxicities, based on data from phase 2 and 3 trials [21-24,33,34] .
Less than 1% to 2% of patients treated with PARPis have also gone on to develop myelodysplastic syndrome
or acute myeloid leukemia (AML), but it had been unclear whether this development was due to exposure
to PARP inhibitor, prior chemotherapy (alkylating agents or anthracyclines), or additive effects of
treatment. The recently published SOLO-1 trial evaluating frontline olaparib maintenance also showed
a 1% incidence of AML in the treatment arm (compared to 0% in the placebo arm), which is worrisome