The first poly ADP ribose polymerase (PARP) inhibitor was developed in the early 1990s. Since then, the activity of PARP inhibitors has been explored in a variety of settings, including and perhaps most enthusiastically in the treatment of cancer. The greater dependence of several cancers on PARP, an important regulator of the DNA base-excision-repair pathway, paved the road for this polymerase as an attractive therapeutic target for cancer.

After nearly 25 years of investigation of the mechanisms of action and efficacy of this class of agents, olaparib (Lynparza) became the first PARP inhibitor in a new class of drugs to be approved for use in cancer treatment in December 2014. The drug is approved for use in women with advanced heavily pretreated ovarian cancer associated with specific abnormalities in the BRCA gene determined by a companion diagnostic called BRACAnalysis CDx which detects BRCA gene mutations in blood samples from patients with ovarian cancer.

Initial Enthusiasm

When preliminary data from a randomized phase II study of the PARP inhibitor iniparib in women with triple-negative breast cancer were presented at the ASCO 2009 Annual Meeting, there was abundant enthusiasm over an emerging novel targeted therapy with a potential indication in this disease setting. Preclinical studies showed that combining PARP inhibitors with platinum chemotherapy, which induces DNA damage through adducts and crosslinking, may potentiate chemotherapy toxicity. This excitement was paired with a growing and strong rationale for treating triple-negative breast cancer with DNA-damaging agents like the platinums, based on the similarities in the gene-expression profiles of BRCA1-deficient and sporadic triple-negative breast cancers.

Thus, in that phase II trial published in 2011,¹ 123 patients were randomized to receive gemcitabine and carboplatin alone or with the addition of iniparib. The study demonstrated an extension of progression-free survival from 3.6 to 5.9 months and a statistically significant overall survival advantage from 7.7 to 12.3 months with the addition of iniparib, with few or no additional side effects. Despite its limitations and the slight imbalance in prognostic factors favoring the iniparib group, the study provided proof of concept.

On the basis of these thrilling results, iniparib was welcomed as the first promising targeted agent for triple-negative breast cancer. The great enthusiasm generated by the fairy tale phase II trial findings in a poor prognostic breast cancer subgroup desperately in need of a success was widely embraced with open arms. The resulting high expectations, despite the underwhelming performance and modest activity of the gemcitabine/carboplatin doublet in this trial, generated great uptake of this combination in triple-negative breast cancer.

Overwhelming Disappointment

The subsequent open-label phase III trial adequately powered to assess the addition of iniparib to gemcitabine-carboplatin accrued rapidly, and an expanded access program to provide triple-negative breast cancer patients with access to iniparib prior to the planned U.S. Food and Drug Administration submission and approval ensued. The outcome of the trial—final results of which were recently reported by O’Shaughnessy and colleagues,² as reviewed in this issue of The ASCO Post—was an overwhelming disappointment; the coprimary endpoints were not met, with iniparib treatment being associated with nonsignificant increases of 1 month in median progression-free survival and 2.5 months in median overall survival.

This trial’s unfortunate results were followed by a phase III trial in squamous cell lung cancer and a phase II trial in platinum-resistant ovarian cancer, which also failed to demonstrate an advantage for iniparib-containing treatment. In June 2013, the manufacturer dropped development of the drug.

What Went Wrong?

So, what went wrong with the phase III trial, particularly following such exciting phase II data? The use of iniparib was clearly biologically plausible, and its evaluation in the triple-negative breast cancer population seemed ideal, since these tumors bore elements and features similar to those of BRCA-mutant tumors. In addition, emerging data in BRCA-associated ovarian cancer suggested a signal of efficacy for this class of drug.

The randomized open-label trial enrolled 519 patients with baseline characteristics that were well balanced, with 57% receiving first-line therapy and 43% receiving second- or third-line therapy. Exploratory subgroup analyses by prior lines of therapy suggested a potential benefit in second- and third-line patients, but it is difficult to generate a hypothesis to explain this further.

Although it is well recognized that triple-negative breast cancers are a heterogeneous group, perhaps the clinical assays used to identify patients for study eligibility were not sufficient for identifying the subgroup most likely to benefit from PARP inhibition. The signal that a given subgroup may benefit and that iniparib might work in certain molecular subtypes of triple-negative breast cancer suggests that perhaps intrinsic subtyping may be of benefit. However, extensive biomarker analyses performed after the initial phase III results came out failed to identify a clear predictive marker. Analysis of the molecular subtyping of specimens from more than 300 patients enrolled in the trial with regard to exposure and response to iniparib may help to further define characteristics associated with benefit or no benefit.

The fact that patients in the control group of this trial were permitted to cross over to iniparib has been implicated as potentially biasing the overall survival results. Overall, 59% of patients in the control arm were allowed to cross over and receive iniparib after central confirmation of disease progression. As usual, allowance of a sequential exposure to iniparib in the control arm, which greatly facilitated rapid accrual to the trial, could reflect receipt of a more active therapy of proven value than the subsequent lines of standard chemotherapy administered beyond disease progression in the investigational arm.

Confounding this further were subsequent reports detailing that the mechanism of action of iniparib was not a true PARP inhibitor; it exhibits 1,000-fold less activity than other PARP inhibitors, a feature attributed to a carboxyl group that potentially weakens its bond to PARP. This difference in mechanism of action may also result in its apparent lack of toxicity seen with other PARP inhibitors, such as the recently approved olaparib, particularly in combination with chemotherapy. The signal of benefit noted in the phase II trial and in the second- and third-line patients in the phase III trial suggests that iniparib probably works differently from what was originally expected and from a true PARP inhibitor.

Challenges Ahead

PARP inhibitors still look to be an effective class of drug, although current evidence indicates that their benefit is largely restricted to BRCA-mutation carriers. The ability to combine these agents with chemotherapy has been and remains challenging. For the future, use of molecular diagnostics may help to predict who may be best suited to benefit from PARP inhibitors.

The iniparib story is a tale of caution and underscores the challenges in the development of novel therapeutics. A number of challenges continue to confront trial designs designated for regulatory agency approvals of targeted agents, so selection of comparators and identification of specialized populations become even more critical. In addition, although a lack of crossover design is poorly received by patients, physicians, and advocates alike, this may have certainly played a role in iniparib’s plummeting descent. Although iniparib fumbled and shattered dreams, the shards left continue to be a stark reminder of the work that remains undone. ■

Disclosure: Dr. Yardley reported no potential conflicts of interest.

References

1. O’Shaughnessy J, Osborne C, Pippen JE, et al: Iniparib plus chemotherapy in metastatic triple-negative breast cancer. N Engl J Med 364:205-214, 2011.

2. O’Shaughnessy J, Schwartzberg L, Danso MA, et al: Phase III study of iniparib plus gemcitabine and carboplatin versus gemcitabine and carboplatin in patients with metastatic triple-negative breast cancer. J Clin Oncol 32:3840-3847, 2014.