Onvansertib

Onvansertib is Cardiff Oncology's oral PLK1 inhibitor. In this Phase 1 trial (NCT05383196), it's being combined with paclitaxel for metastatic triple-negative breast cancer at Mayo Clinic, under investigator-sponsorship from Antonio Giordano MD [4]. The study is dose-finding with a planned expansion, n=50, currently active and not recruiting [4]. The TNBC program sits outside Cardiff's main commercial bet, which is a Phase 2 trial in first-line KRAS/NRAS-mutant metastatic colorectal cancer (CRDF-004, NCT06106308), where onvansertib's earlier Phase 1b/2 data showed activity worth pursuing [1,2,5]. For Cardiff, this TNBC trial is a low-cost option on a second indication. For patients with TNBC, it adds one more hypothesis to a space already reshaped by sacituzumab govitecan in chemo-pretreated disease, pembrolizumab plus chemotherapy in PD-L1-positive front-line, and PARP inhibitors (olaparib, talazoparib) in germline BRCA-mutant disease. The drug is not approved anywhere, has no FDA designations specific to breast cancer, and the trial uses no biomarker selection.

Onvansertib is a novel compound, not approved in any indication anywhere [4]. It originated at Nerviano Medical Sciences (as NMS-1286937/PCM-075), was licensed to Trovagene (now Cardiff Oncology, ticker CRDF), and is the company's lead asset. The drug is selective for PLK1 over PLK2 and PLK3 - a distinction that matters because the pan-PLK inhibitor volasertib failed its Phase 3 VALOR trial in AML on both efficacy and an early mortality imbalance [8]. The TNBC Phase 1 (NCT05383196) is investigator-sponsored, active-not-recruiting, with no FDA designations attached [4]. As of this writeup, no interim or topline efficacy data from NCT05383196 has been disclosed in peer-reviewed publications, conference abstracts, or Cardiff press releases that we can locate; readers should treat the signal as unknown until presented. The most material near-term catalyst for the drug as a whole is data from CRDF-004 (NCT06106308), a Phase 2 trial of onvansertib added to FOLFIRI/bevacizumab or FOLFOX/bevacizumab as first-line in KRAS/NRAS-mutant mCRC, now active-not-recruiting at n=113 [5]. A completed Phase 2 in pancreatic ductal adenocarcinoma (NCT04752696, onvansertib + nanoliposomal irinotecan + 5-FU/leucovorin) is also on the docket for data [6]. The TNBC trial sits on the periphery. An investigator-sponsored Phase 1 with a 50-patient ceiling is not the kind of study a sponsor builds a registration package around. Expected readout for the TNBC study is not formally announced, but active-not-recruiting status suggests data presentation in H2 2026 or 2027, likely at SABCS or ASCO.

PLK1 (polo-like kinase 1) is a regulator of mitosis, the chaotic phase where a cell duplicates itself. Think of PLK1 as a foreman during that phase: it activates checkpoints, organizes the mitotic spindle (the apparatus that pulls duplicated chromosomes apart), and times the exit from mitosis. Cancer cells often overexpress PLK1 and depend on it more than normal cells do, because they divide faster and accumulate more genomic stress [1,2]. Blocking PLK1 traps cells in mitosis, which then triggers apoptosis (programmed cell death). Onvansertib is an ATP-competitive inhibitor with reported biochemical IC50 of ~2 nM against PLK1 and >5,000 nM against PLK2 and PLK3 - a >1000-fold selectivity window in cell-free assays, per the Nerviano preclinical characterization [9]. That contrasts sharply with volasertib, which inhibits PLK1, PLK2, and PLK3 with broadly similar potency and ran into a fatal toxicity signal in late-stage AML trials [8]. The paclitaxel pairing is mechanistically clean: paclitaxel stabilizes microtubules and stalls cells at the spindle assembly checkpoint, exactly where PLK1 activity matters most. Two drugs, one vulnerable phase, in principle. Validation for PLK1 as an oncology target is mixed. The genetic rationale is real, since PLK1 overexpression correlates with worse outcomes in multiple tumor types. But the clinical track record is rough: no approved PLK1 inhibitor, and prior compounds have stumbled on safety or modest single-agent activity. Onvansertib's KRAS-mutant CRC signal from the Phase 1b/2 study is the strongest clinical evidence the class has produced to date [1,2].

NCT05383196 is a Phase 1 single-arm trial with a dose-limiting toxicity (DLT) primary endpoint, meaning the first question is what dose of onvansertib can be safely added to standard paclitaxel in this population [4]. Enrollment target is 50; the trial is now active-not-recruiting, which suggests accrual is complete and the team is in follow-up. The sponsor is Antonio Giordano MD at Mayo Clinic, not Cardiff Oncology directly - this is an investigator-initiated study, the kind a company supports with drug supply and modest funding but doesn't bet a registration on. The population is metastatic TNBC, with no biomarker selection (no requirement for PLK1 expression, BRCA status, or any other enrichment marker) [4]. The unselected design has a wrinkle: roughly 15-20% of TNBC patients carry germline BRCA1/2 mutations and now have approved PARP inhibitor options (olaparib, talazoparib) in the metastatic setting, so the enrolled population is a mix of BRCA-mutant and wild-type patients with materially different standards of care. Any expansion signal will need a post-hoc cut by BRCA status to be interpretable. There's no comparator arm, which is standard for a dose-finding study but makes efficacy interpretation in the expansion cohort tricky. Paclitaxel alone has activity in TNBC, so any response could be carried by the backbone. Endpoints beyond DLT include response rate and duration of response. The design is appropriate for what it is: a hypothesis-generating bet that PLK1 inhibition adds something on top of taxane chemotherapy. Whether that hypothesis is worth pursuing in a randomized Phase 2 depends entirely on what the expansion data look like.

Our model estimates a 6% chance this drug is eventually approved. It starts from the historical base rate for Phase 2 drugs in this area (about 13%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by a non-randomized design and its light or open-label blinding; it is held back by the sponsor's thin or weak approval record and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is real. There is no biomarker selection in this trial. Patients aren't enriched for PLK1 dependency, BRCA status, or any other predictive marker [4]. Single-arm against a backbone (paclitaxel) with known activity means any signal is hard to attribute. TNBC is heterogeneous, and previous attempts to layer novel agents on taxanes without enrichment have mostly failed to outperform the backbone. The unselected design also means germline BRCA-mutant patients, who have approved PARP inhibitor alternatives (olaparib, talazoparib), are in the same cohort as wild-type patients with worse standard-of-care options - any TNBC Phase 2 hypothesis will need to address whether the signal is BRCA-status dependent. Safety risk: PLK1 inhibition causes neutropenia and thrombocytopenia mechanistically, and paclitaxel does the same. Overlapping myelosuppression is the most likely DLT-defining toxicity. Volasertib's VALOR Phase 3 in AML showed an early mortality imbalance that ended the program [8], and while onvansertib's >1000-fold selectivity for PLK1 over PLK2/3 [9] is a different drug profile, the class memory is hostile. Execution risk: this is an investigator-sponsored Phase 1 with a 50-patient ceiling at a single institution. Enrollment pace, data quality, and publication speed all depend on academic incentives rather than corporate ones. Commercial risk: TNBC is already served by sacituzumab govitecan in chemo-pretreated patients, pembrolizumab plus chemotherapy in PD-L1-positive front-line, PARP inhibitors in germline BRCA-mutant disease, and datopotamab deruxtecan (now approved in HR+/HER2- metastatic breast cancer per TROPION-Breast01, with TNBC indication still investigational pending TROPION-Breast02). A Phase 1 combination with paclitaxel needs to clear a high bar to differentiate, even if biological signal is real. Cardiff Oncology itself is small-cap (CRDF) with limited cash runway - public filings have historically shown cash positions in the $30-50M range against operating burn of roughly $10M per quarter, implying a runway measured in single-digit quarters absent additional financing; readers should check the most recent 10-Q for current numbers before sizing TNBC follow-on bets.

Noise for now. The TNBC trial is exploration on Cardiff Oncology's secondary front. The company's stock-moving catalyst is somewhere else: CRDF-004, the Phase 2 in first-line KRAS/NRAS-mutant metastatic colorectal cancer (NCT06106308) [5]. That's where the validation question gets answered. Does the KRAS-mutant CRC signal from the Phase 1b/2 study [1,2] replicate in a randomized setting? The TNBC Phase 1 is option value, not a thesis. A signal that would change the read on the TNBC program: response rate in the expansion cohort meaningfully above paclitaxel alone (above 30-35% ORR), with a tolerable safety profile, no surprise toxicities, and ideally a BRCA-stratified cut showing activity in wild-type patients (where competition from PARP inhibitors is absent). Anything less is publishable but not commercially interesting. Check back when NCT05383196 reports, likely at SABCS or ASCO presentations in late 2026 or 2027 given the active-not-recruiting status. The faster way to track Cardiff's trajectory is CRDF earnings releases and 8-K filings around CRDF-004 readouts, which will determine whether the company has a real asset or a failed thesis. For BioCosm purposes, this node is low priority until expansion data drops.