USP1 Coleman: AOH1996 and The New Pan-Cancer Treatment

AOH1996 and USP1 inhibitors represent two of the most promising experimental strategies to emerge from cancer research in recent years, both built on the idea of exploiting weaknesses in how cancer cells copy and repair their DNA. AOH1996 is a small molecule developed at City of Hope and named after a pediatric cancer patient born in 1996. It targets a cancer-specific form of a protein called PCNA, and preclinical testing has shown activity across more than 70 different solid-tumor cell lines with minimal harm to normal cells. USP1 inhibitors take a related but distinct approach, disabling an enzyme cancer cells depend on to manage DNA damage. Together, these strategies point toward a shift in how researchers think about treating cancer: not by poisoning all fast-dividing cells, but by targeting the molecular machinery that tumors rely on to survive their own genetic chaos.

The Discovery That Started It All

The story behind AOH1996 begins with an observation made about the protein PCNA, or proliferating cell nuclear antigen. PCNA is sometimes called a “sliding clamp” because it encircles DNA and acts as a platform for dozens of proteins involved in copying and repairing the genome. Every dividing cell uses PCNA, which is why researchers initially assumed it would be a poor drug target: blocking it would harm healthy tissue just as much as cancerous tissue.

That assumption changed when researchers at City of Hope discovered that cancer cells carry a structurally different version of PCNA. This cancer-associated isoform, which they called caPCNA, was found in breast cancer cells and tumor tissues but was largely absent from non-malignant cells.¹ Specifically, the team found that a particular region of the protein (residues L126 through Y133) was structurally altered in cancer cells, making it more accessible to interactions with other proteins.² The difference was subtle enough that standard laboratory antibodies could not distinguish between normal PCNA and caPCNA, but it was real, and it opened the door to a drug that could tell the two apart.

How AOH1996 Kills Cancer Cells

Using a rational drug design approach, the City of Hope team identified a small molecule that specifically targets caPCNA. They named it AOH1996. The compound works by binding at the interface where PCNA trimers interact with each other, locking the protein into an inactive shape that prevents it from doing its job.³ Crystal structures show that three molecules of AOH1996 wedge themselves between two PCNA trimers, effectively jamming the machinery.

What happens next is where things get interesting. Cancer cells are constantly copying their DNA at a frantic pace, and at the same time they are reading genes to produce proteins. When these two processes collide on the same stretch of DNA, you get what researchers call a transcription-replication conflict. In normal cells, PCNA helps manage these collisions. In cancer cells treated with AOH1996, the drug strips PCNA away from actively transcribed regions of the genome, causing replication forks to collapse specifically where genes are being read. The result is an accumulation of double-stranded DNA breaks, which are among the most lethal forms of DNA damage a cell can sustain.⁴

The key detail is that this damage is transcription-dependent. When researchers used a chemical that shuts down transcription entirely, AOH1996 stopped causing DNA damage. The drug also triggered the accumulation of R-loops, unusual three-stranded structures made of DNA and RNA that form when transcription and replication collide.⁴ In quiet, non-transcribed portions of the genome, PCNA stayed put and replication continued undisturbed. This selectivity for highly active chromatin regions is part of why AOH1996 preferentially damages cancer cells, which tend to be both replicating faster and transcribing more aggressively than normal tissue.

Why Normal Cells Are Largely Spared

The selective toxicity of AOH1996 has been one of its most striking features in preclinical testing. Agents that target caPCNA can kill cancer cells without causing significant toxicity to normal cells or experimental animals.⁵ Testing of AOH1996 analogs in a non-cancerous cell line confirmed this pattern, showing reduced cytotoxic effects compared to what was seen in tumor cells.⁶

This selectivity likely comes from several converging factors. Normal cells express standard PCNA, not the caPCNA variant, so AOH1996 has a weaker grip on its target in healthy tissue. Normal cells also have lower rates of both replication and transcription, meaning fewer opportunities for the transcription-replication collisions that AOH1996 exploits. And importantly, normal cells have intact DNA-repair pathways that can handle moderate levels of replication stress, while many cancers have already disabled parts of their repair machinery through accumulated mutations. The drug essentially capitalizes on vulnerabilities that cancer cells have created for themselves.

Preclinical Results Across Cancer Types

The “pan-cancer” label attached to AOH1996 comes from the breadth of tumor types it has shown activity against. Preclinical data demonstrate activity in more than 70 solid-tumor cell lines, including breast, lung, colorectal, and prostate cancers, with negligible effects on normal cells.³ This makes sense given that caPCNA appears to be a near-universal feature of malignant cells rather than something specific to one tissue type.

More recent studies have started drilling into individual cancer types. In pancreatic ductal adenocarcinoma, which is among the deadliest cancers, AOH1996 showed dose-dependent killing that was tied to the activity of the KRAS oncogene. Organoids with the highest replication stress were the most sensitive to the drug. In mouse models, AOH1996 reduced tumor growth and extended median survival from 14 days to 21 days without apparent toxicity.⁷

In head and neck squamous cell carcinoma, AOH1996 inhibited tumor growth and invasion both in lab dishes and in living animals. It also suppressed cancer stemness, the ability of tumor cells to behave like stem cells and regenerate the tumor after treatment. Perhaps most intriguingly, AOH1996 shifted the tumor’s immune environment toward an inflamed state with more CD8+ T cells infiltrating the tumor, suggesting potential synergy with immunotherapy.⁸

What USP1 Does and Why It Matters

To understand USP1 inhibitors and how they fit alongside AOH1996, you need to know a bit about how cells manage DNA damage during replication. When a replication fork runs into a damaged stretch of DNA it cannot copy, the cell tags PCNA with a small protein called ubiquitin. This tag acts as a signal to swap in specialized “sloppy” DNA polymerases that can copy past the damage, a process called translesion synthesis.⁹ After UV damage, for example, PCNA gets tagged with a single ubiquitin molecule, and this modified form specifically recruits the bypass polymerases needed to get past the lesion.¹⁰

USP1 is the enzyme responsible for removing that ubiquitin tag once the crisis has passed. It acts as a molecular reset button, stripping ubiquitin from PCNA so the normal replication machinery can take over again. USP1 also deubiquitinates another protein called FANCD2, which is central to repairing a particularly dangerous type of DNA damage known as interstrand crosslinks.¹¹ Working with a partner protein called UAF1, USP1 keeps these repair signals tightly regulated.¹²

USP1 Inhibitors and Synthetic Lethality

The therapeutic potential of blocking USP1 comes from a concept called synthetic lethality. The idea is straightforward: cancer cells that have already lost one DNA-repair pathway become fatally dependent on the remaining ones. Knock out a second pathway, and the cell cannot cope. This is the same logic behind PARP inhibitors, which have been transformative for patients with BRCA1/2-mutant cancers. USP1 inhibitors open a parallel avenue of attack.

Research has shown that eliminating USP1 function is synthetically lethal in cancers carrying BRCA1/2 mutations, where the homologous recombination repair pathway is already broken.¹³ When USP1 is inhibited in these tumors, ubiquitinated PCNA accumulates unchecked, flooding the cell with error-prone translesion synthesis and destabilizing the genome beyond rescue.¹⁴ The effect is not limited to BRCA-mutant tumors; a subset of BRCA-wild-type tumors with other forms of homologous recombination deficiency are also vulnerable.

One important nuance: the synthetic lethality between USP1 inhibition and homologous recombination deficiency is not fully penetrant. In testing of the USP1 inhibitor TNG348, the six most sensitive cell lines were all BRCA1/2-mutant and HRD-positive, but seven other cell lines with the same genetic profile showed lower sensitivity.¹⁵ This means the BRCA mutation alone does not guarantee a response, and finding the right biomarkers to predict which patients will benefit is an active area of investigation.

Overcoming PARP Inhibitor Resistance

One of the most clinically urgent problems USP1 inhibitors could address is resistance to PARP inhibitors. PARP inhibitors like olaparib have been a genuine breakthrough for patients with BRCA-mutant ovarian, breast, and other cancers, but many tumors eventually develop resistance and start growing again. For those patients, options have been limited.

KSQ-4279, the first USP1 inhibitor to reach clinical testing, was developed specifically with this problem in mind. In preclinical models, the combination of KSQ-4279 with a PARP inhibitor was well tolerated and induced durable tumor regression across several patient-derived models that had already become resistant to PARP inhibitors alone.¹⁶ Structural studies have revealed how KSQ-4279 binds within a hydrophobic tunnel in the USP1 protein, stabilizing its fold and locking it in an inactive state.¹⁷

Separate work has uncovered a direct biological link between USP1 and PARP1: USP1 removes ubiquitin from PARP1 itself, regulating how PARP1 gets trapped on DNA and how much of its signaling activity it retains. When both USP1 and PARP1 are blocked simultaneously, the resulting replication stress and DNA damage spike, killing even cells that had become resistant to either platinum-based chemotherapy or PARP inhibitors. This effect was observed regardless of whether the cancer cells had intact homologous recombination, which is particularly encouraging.¹⁸

AOH1996 and Combination Therapies

AOH1996 is not just being studied as a standalone agent. Because it interferes with PCNA’s role in managing replication stress, it can sensitize cancer cells to other DNA-damaging treatments. In preclinical testing, AOH1996 increased the sensitivity of cancer cells to cisplatin, a standard platinum-based chemotherapy drug, and to topotecan, a topoisomerase I inhibitor. Cisplatin primarily creates bulky DNA adducts in active regions of the genome, and topotecan prevents a cutting enzyme from resealing DNA after it makes nicks, leading to strand breaks during replication. AOH1996 amplifies both of these effects by disabling the PCNA-mediated repair response that cancer cells would normally mount.⁴

The finding that AOH1996 promotes immune-cell infiltration in head and neck cancer models also raises the prospect of combining it with checkpoint immunotherapy. When AOH1996 was paired with anti-PD1 treatment in squamous cell carcinoma, the combination outperformed either treatment alone, with the inflamed tumor microenvironment and increased CD8+ T-cell presence creating a setting where immunotherapy could gain a foothold.⁸ This aligns with a broader pattern in the field: drugs that increase replication stress and DNA damage in tumors often make those tumors more visible to the immune system, because broken DNA fragments leak into the cell’s cytoplasm and trigger innate immune signaling.

The Broader Push to Target Replication Stress

AOH1996 and USP1 inhibitors are part of a larger wave of drugs designed to exploit a fundamental vulnerability of cancer: replication stress. Because tumors accumulate mutations, lose checkpoint controls, and drive their cells to divide relentlessly, they operate under constant strain during DNA copying. This is not a minor inconvenience for the tumor. It is a survival bottleneck that cancer cells manage only by leaning hard on a handful of repair and checkpoint proteins.¹⁹

Other drugs in this space target kinases like ATR, CHK1, WEE1, and MYT1, which coordinate the cell’s response to replication problems. What sets AOH1996 apart is that it strikes at the replication machinery itself rather than at the signaling network around it. And what sets USP1 inhibitors apart is their specificity for cancers that have already compromised homologous recombination repair, making them a natural successor or complement to PARP inhibitors. The two approaches are not competitors. They exploit different aspects of the same underlying problem and may eventually be combined, either with each other or with existing therapies, to attack tumors from multiple angles simultaneously.

Drugging a Target That Was Not Supposed to Be Druggable

One underappreciated piece of the AOH1996 story is that PCNA was long considered undruggable. It is a flat, ring-shaped protein with no obvious pocket where a small molecule could bind. The breakthrough came from discovering that PCNA has a cryptic binding site, one that does not exist in the protein’s resting state but forms transiently when the protein shifts between structural states. Fluorescence measurements showed that the core of PCNA is extremely nonpolar, and mutagenesis studies localized a hidden binding pocket near residues L90 and L101. Ligands that bind to this transient site stabilize the PCNA trimer and reduce its ability to disassemble and reassemble its subunits, which is exactly how AOH1996 locks PCNA into an inactive conformation.²⁰

This matters beyond just AOH1996 because it demonstrates that proteins written off as undruggable may simply have binding sites that are invisible to standard structural analyses. The discovery of cryptic sites, pockets that open only fleetingly and require dynamic methods to detect, is reshaping how medicinal chemists think about target selection. For PCNA specifically, the finding that AOH1996 wedges between trimers at a site that only appears under certain conditions explains why decades of drug-development efforts failed before the structural biology caught up with the biology.

Where These Drugs Stand Clinically

AOH1996 entered a Phase I clinical trial for advanced solid tumors (registered as NCT05227326), marking the first time a drug specifically targeting caPCNA has been tested in humans. Phase I trials focus primarily on safety and dosing, so efficacy data are still emerging. The preclinical profile, activity across dozens of tumor types with low toxicity, is promising but the gap between mouse models and human patients has humbled many cancer drugs before.

For USP1 inhibitors, KSQ-4279 has advanced furthest into clinical development. The preclinical case for combining it with PARP inhibitors in BRCA-mutant, PARP-resistant cancers is strong, and clinical trials are testing whether that translates to real patient benefit.¹⁶ Other USP1 inhibitors, including TNG348, are also in development, each with slightly different selectivity profiles and binding mechanisms.¹⁵

The incomplete penetrance of USP1 synthetic lethality, where some BRCA-mutant tumors respond strongly and others with the same mutation do not, means that companion diagnostics and biomarker strategies will be central to the clinical program. Identifying which patients are most likely to benefit before treatment begins is as important as the drug itself. The broader replication-stress field faces a similar challenge: these drugs work by exploiting a vulnerability that exists on a spectrum, and where a particular tumor falls on that spectrum determines whether the treatment will be transformative or merely marginal.

Replication Stress and the Immune System

An unexpected dimension of replication-stress drugs is their capacity to make tumors more responsive to immunotherapy. When AOH1996 triggers double-stranded DNA breaks and replication-fork collapse in cancer cells, some of that damaged DNA escapes into the cytoplasm. Cytoplasmic DNA is normally a danger signal, something cells interpret as evidence of viral infection, and it activates innate immune pathways that produce inflammatory signals. In a tumor, this inflammation can recruit immune cells, particularly cytotoxic T cells, and make the tumor more susceptible to checkpoint blockade.

The head and neck cancer study demonstrated this directly: AOH1996 shifted the tumor immune microenvironment toward an inflamed state with increased CD8+ T-cell infiltration, creating favorable conditions for anti-PD1 immunotherapy.⁸ Whether USP1 inhibitors produce similar immunogenic effects has not been studied as thoroughly, but the shared mechanism of amplified replication stress and DNA damage suggests the possibility. If replication-stress drugs consistently warm up cold tumors, their value may extend well beyond the specific repair pathways they target, turning them into general-purpose partners for the immunotherapy drugs that have already transformed cancer treatment for some patients but continue to fail for many others with immunologically quiet tumors.

1
PubMed Central. A cancer-associated PCNA expressed in breast cancer has implications as a potential biomarker
2
PubMed Central. The anti-cancer activity of a first-in-class small molecule targeting PCNA
3
PubMed Central. Small molecule targeting of transcription-replication conflict for selective chemotherapy
4
Cell Chemical Biology. Small molecule targeting of transcription-replication conflict for selective chemotherapy
5
PubMed Central. Therapeutic Targeting of DNA Replication Stress in Cancer
6
PubMed Central. Synthesis and Anticancer Evaluation of PCNA Inhibitor AOH1996 Analogs in Cancer Cell Cultures
7
PubMed. Therapeutic Targeting of Oncogene-Induced Transcription-Replication Conflicts in Pancreatic Ductal Adenocarcinoma
8
PubMed Central. The PCNA inhibitor AOH1996 suppresses cancer stemness and enhances anti-PD1 immunotherapy in squamous cell carcinoma
9
PubMed Central. Structure of monoubiquitinated PCNA and implications for translesion synthesis and DNA polymerase exchange
10
Molecular Cell. Monoubiquitination of PCNA and Its Role in Interacting with Human DNA Polymerase η
11
Molecular Cell. The Deubiquitinating Enzyme USP1 Regulates the Fanconi Anemia Pathway
12
PubMed. USP1 in regulation of DNA repair pathways
13
PubMed Central. USP1 inhibition: A journey from target discovery to clinical translation
14
PubMed Central. Ubiquitinated PCNA Drives USP1 Synthetic Lethality in Cancer
15
Molecular Cancer Therapeutics. Characterization of TNG348: A Selective, Allosteric USP1 Inhibitor That Synergizes with PARP Inhibitors in Tumors with Homologous Recombination Deficiency
16
PubMed Central. The USP1 Inhibitor KSQ-4279 Overcomes PARP Inhibitor Resistance in Homologous Recombination-Deficient Tumors
17
PubMed Central. Structural and Biochemical Insights into the Mechanism of Action of the Clinical USP1 Inhibitor, KSQ-4279
18
PubMed Central. USP1 deubiquitinates PARP1 to regulate its trapping and PARylation activity
19
PubMed Central. Targeting replication stress in cancer therapy
20
Cell Press (Biophysical Journal). Discovery of a novel cryptic binding site in the hydrophobic core of human proliferating cell nuclear antigen