A genome-wide CRISPR screen in Brca2‑deficient murine prostate organoids identifies loss-of-function in DNA prereplication complex genes (CDT1, CDC6, DBF4) as a reversion‑independent mechanism of resistance to PARP inhibitors; pharmacologic disruption of the Geminin–CDT1 interaction can restore sensitivity.
0:00Welcome to Base by Base, the paper cast that brings genomics to you wherever you are. Thanks for listening, and don't forget to follow and rate us in your podcast app. Imagine you have a highly advanced targeted weapon designed to exploit a specific flaw in a cancer cell's armor.
0:14Right, like a therapy that is tailor made for a tumor's exact genetic profile. Exactly. The clinical data tells you it should be a total slam dunk, you know? For half the patients, it works perfectly. The cancer cells simply cannot withstand the attack and they die.
0:29But for the other half of the patients, the cancer mysteriously survives. I mean, it is arguably one of the most frustrating phenomena in oncology today. The drug binds to the intended target, and the genomic profile of the tumor dictates that the cell should undergo epoctosis.
0:45Yeah, by all accounts, it should be a lethal hit. Right. Yeah, the clinical income diverges entirely from the molecular theory. The cells just keep dividing. It really prompts you to wonder, what really happens when a cancer cell takes a vital piece of its own replication machinery and, well, breaks it just enough to survive a deadly drug.
1:02And more importantly, how could this change the way we predict which patients will actually respond to targeted therapy? It's almost like the cancer cells are on a sinking ship. You plug the main hole, and by all laws of physics, they should sink.
1:16But somehow these cells are building a completely new buoyancy system on the fly. Today we celebrate the work of Kyrie Papas, Charles L. Sawyers, and the research team, and Memorial Sloan Kittering Cancer Center, who have advanced our understanding of PRP inhibitor resistance in prostate cancer.
1:33And just to give you some timeline context for this deep dive, This research was published in the journal PNAS on June 3, 2025. It's really incredible piece of work. It is. But to understand why the MSKCC team needed to investigate this in the 1st place, we really have to look at the clinical reality of metastatic castration resistant prostate cancer or MCRPC.
1:53The landscape for MCRPC has shifted significantly over the last decade. Historically, once a tumor bypassed androgen deprivation therapy, the options were severely limited. Right. You were kind of out of roadmap at that point.
2:06Exactly. But the introduction of PRP inhibitors, like OlaParib, change that trajectory entirely. This is specifically a breakthrough for tumors harboring BRCA1 or BRCA2 mutations. Now most of the time, when you hear BRCA mutations, your mind immediately goes to breast and ovarian cancer.
2:25And typically, as a germ line mutation that someone inherited from a parent. Yes, that's the classic association most people have. But in the context of prostate cancer, the genomic profile is distinctly different, right?
2:36BRCA 2 mutations heavily outnumber BRCA one, and more importantly, roughly half of these BRCA mutations are somatic. Which means the patient wasn't born with the defect. The tumor actually acquired it spontaneously during its evolution.
2:51And because these mutations completely abrogate homologous recombination, the cell loses its primary mechanism for error-free repair of DNA, double tram brakes. So it's incredibly vulnerable. Very. PRP inhibitors are engineered to exploit that very specific synthetic lethality.
3:06You knock out the backup repair pathway with the drug, and a BRCA deficient cell basically has no way to fix its DNA. But let's pause on the clinical reality for a 2nd because this is the entire reason the study exists.
3:17About 50% of these BRCA mutant prostate cancer patients exhibit upfront resistance to PRP inhibitors. Half of the patients. I mean, they have the precise biomark that the drug was built for, yet they derive 0 clinical benefit from day one.
3:32Wow. Yeah, the tumor just shrugs off the therapy as if the homologous recombination deficiency wasn't even there. Okay, let's unpack this. If PRP inhibitors work by trapping the PRP1 protein directly on the DNA strand to cause a, like a fatal traffic jam for the cellular replication machinery.
3:51How are these BRCA deficient cells surviving the crash? That is the big question. Because they are already completely compromised in their DNA repair pathways. A replication for colliding with a trapped PRP1 complex should be a catastrophic terminal event for that cell.
4:07And that collision is usually lethal. When the replicum hits that trapped PRP1 protein, the replication fork stalls out. Without BRCA mediated repair to stabilize and restart that fork, it collapses into a highly toxic double strand break.
4:21If a BRCA deficient tumor is surviving that onslaught, it implies a fundamental rewiring of how the cell manages replication stress. Like, before the fork ever has a chance to collapse in the 1st place.
4:33Exactly. It's managing the stress upstream. So to untangle that rewiring, the researchers needed a robust model. But you can't just pull standard human MCRPC cell lines off the shelf for this, can you?
4:45No, you really can't. Previous attempts using engineered human cell lines, like LNTP, they only yielded modest sensitivity to PRP inhibitors. Which doesn't really help. Right, because it doesn't accurately reflect the profound synthetic legality we observe in actual human patients.
5:01The standard flat dish cell lines lack the complex struml interactions, and the specific genetic background required to truly model this disease. So the team pivoted to primary mouse prostate organoids.
5:12And these 3D cultures maintain the crucial epithelial architecture, which is so important. They do. They took these wild type organoids and used CRISPR to introduce a bioliolic loss of Braki too. But they didn't stop there.
5:24They also did targeted knockouts of TTP 53, 10, and RB1. They didn't just knock out the primary target. They layered in the loss of major tumor suppressors to mirror the aggressive heterogeneous genetic landscape you actually see in a patient's tumor.
5:42Yeah, that multi-hit genetic background was the key. By establishing that specific environment. The organoid suddenly exhibited the exquisite sensitivity scene in the clinic. We are talking about the IC 50.
5:54That's the concentration required to inhibit growth by 50% plummeting, right? Oh, absolutely plummeting. It dropped from the 10 micromolar range all the way down to between 10 and 333 nanomolar. Wow. The cells became up to a 1000 times more sensitive to Ola Prib.
6:09Wait, so instead of trying to find the resistance in human cells, They deliberately engineered mouse cells to be incredibly fragile, just so they could hit them with a massive genetic hammer and see what survived.
6:18I know it sounds wild, but the logic of the experimental design demands that fragility. Yeah, if you want to run a genome wide CRISPR screen to uncover resistance mechanisms, your baseline model has to be overwhelmingly susceptible to the drug.
6:31You need the signal to noise ratio to be perfect. Oh I see. That makes a lot of sense. So they utilize the Brie lentiviral library, which contains over 78,000 guide RNAs to systematically knock out genes across 1000000s of these hypersensitive organoid cells.
6:48Exactly. And then they flood the system with a lethal 100 anomolar ghost of OlaParib and just wait for 21 days, which is roughly 10 cell division cycles. And given how fragile they made this model, the vast majority of those cells undergo apoptosis and die.
7:03They do. So the ones that remain alive after 3 weeks, they hold the answer. Right. The genetic knockout they receive from the Lenoviral library must be the mechanical driver of their survival. So when the team sequenced the DNA of the surviving organoids to identify the enriched guide RNAs, the data clustered heavily around 3 specific genes.
7:19CD21, CDC 6, and DBF4. Right, and those 3 are core components of the DNA pre-replication complex, or pre-RC. Now, for those who might need a quick refresher, what exactly is the PreRC doing normally? Well, before a cell can enter S phase and divide, it has to license its replication origins.
7:38Like giving permission to start copying. Exactly. The PreRC physically loads the MCM helicase onto the DNA, effectively marking the exact spots where DNA unwinding and replication will begin. It ensures the genome is copied exactly one's per cell cycle.
7:55Here's where it gets really interesting. It's like the pre-RC is the starter pistol for a race. If the pistol misfires, the runners go slower, but somehow that slower pace prevents them from tripping over the hurdles left by the PRP inhibitor.
8:09That is a great analogy. That altered dynamic is what saves the cell. When you deplete CDT one or CDC 6, you restrict the number of licensed origins. So there's fewer starting lines. Right. Right. The cellfire's fewer replication forks and the overall speed of DNA synthesis drops.
8:24But dropping the speed of replication doesn't remove the trapped PRP one proteins from the DNA. The hurdles are still in the track. Why does slowing down prevent the fatal crash? Comes down to replication fork stability, and um, nucleicide pool management.
8:39Okay. When a cell with normal origin licensing gets hit with a PRP inhibitor, 1000s of replication forks stall simultaneously at those craft proteins. So it's a massive system wide traffic jam. Exactly.
8:52The cell panics and fires dormant origins to train rescue the replication process. But this massive uncoordinated firing exhausts the cellular pool of nucleotides and completely depletes replication factors like RPA.
9:06The cell essentially starves its own repair machinery by trying to fix everything at once, causing the forks to collapse into double strand brakes. Right. By crippling the PreRC, the cancer cell prevents that systemic panic.
9:17With fewer origins firing, the nucleotide pools aren't exhausted. So when a fork stalls at a trapped PRP1 protein. It has the time and the biochemical resources to engage alternative tolerance mechanisms, things like fork reversal or translesion synthesis.
9:32Which allows it to bypass the lesion without the fork collapsing. And the researchers proved this chemically, didn't they? They looked at Gamma H2AX foci, which are the classic cellular markers for double strand DNA breaks.
9:44They did. In the highly sensitive organoids, adding oldaparib caused a massive, sustained spike in those damage markers. I mean, the genomes were shattering. Wow. But in the organoids with the CDT1 or CDC 6 knockouts.
9:58Those Gamma H2AX foci appeared briefly and then resolved within 4 hours. The cells were managing the PREP induced damage on the fly. That's incredible. And they actually map this physically using DNA fiber assays.
10:11If you feed the cells fluorescent thyme day and analogs like CLDU and IDU. You can track the actual progression of individual replication forks under a microscope. Yeah, the fiber essays confirm the mechanism visually.
10:23It's really striking data. What did it look like? Well, in a typical BRCA 2 deficient cell exposed to a PRP inhibitor, the masin DNA strands installed forks are actively degraded by nucleuses. You can literally see the fluorescent tracks getting chewed away.
10:37But when the pre-RC is impaired, those tracks remain completely intact. The slower origin licensing miraculously shields the replication fork from nucleolitic degradation. Hold on. If we look at something like the cancer dependency map, you know, dep map genes like CDT1 are classified as core essential, if you knock out a core component of the pre-replication complex in a standard cancer cell line, it undergoes cell cycle arrest and dies, how is this tumor surviving, let alone forming a resistant shield without a functional pre-RC?
11:08It is a phenomenal display of context dependent tumor biology. In a healthy cell or even mini standard cancer lines, Severe pre RC impairment is lethal. Yeah. But the specific genomic chaos of MCRPC, particularly the concurrent loss of RB1 and TRP 53 allows these cells to tolerate a drastic reduction in origin licensing.
11:29So the other mutations cover for it. Exactly. And to prove this wasn't just an in vitro artifact, the team injected these CDT1 depleted organoids into mice. You'd assume they wouldn't engraft well, given how vital that gene normally is.
11:41You would think so, but they formed robust, subcutaneous tumors. They grew slightly slower than the parental lines, taking a minor hit to their proliferative capacity. But they completely retained the CDT one deletion, and their profound resistance to Olaparib.
11:56We need to bridge this back to the human patient, because a mouse model is only as good as its clinical translation. Does this highly specific pre-RC loss actually happen in real people? The clinical data integration in this paper is arguably its most sobering aspect.
12:12When the team analyzed large scale genomic datasets from patients with castration resistant prostate cancer, they found that roughly 50% of these tumors harbor a copy number loss of these pre-RC complex genes.
12:25Let that sit for a moment. Half of the patients walking into the clinic already possess this underlying genetic alteration. Yeah, for CDT1 specifically, the copy number loss is present in up to 65% of cases, and crucially, these pre-RC losses significantly co-occur with BRCA 2 mutations.
12:41So this isn't some rare acquired resistance mechanism that develops after months of treatment. This is a massive pre-existing shield. The tumors already have this slow replication architecture in place before the PIC inhibitor even enters the patient's bloodstream.
12:56That perfectly explains the 50% upfront failure rate we talked about earlier. The tumor has proactively engineered its escape route. But because the researchers map the exact biochemical pathways of that escape route.
13:09They uncovered a vulnerability to exploit. Right. If the cancer is hiding behind a broken starter pistol. How do we force the race to happen? How do we strip away that shield? We target the regulatory network governing the pre-RC?
13:22Specifically a protein called Gemini? Okay, in a normal cell cycle, Geminin acts as the strict negative regulator of CDT1. During the S and G 2 phases, Geminin physically binds to CDT1 and sequesters it.
13:36It locks CDT1 away to ensure that once a replication origin has fired, it cannot be licensed again. Re-replicating the same segment of DNA is catastrophic for genomic stability. Right. So Gemma is basically the master break on the origin licensing system.
13:49So the cancer survives the 1st drug by slowing down its replication engine. We counterattack by cutting the brakes, meaning gem and forcing the engine to rev out of control until it blows up. Biologically, that is exactly what happens.
14:02The team took the resistant organoids, the ones surviving parp inhibition due to CDC 6 depletion, and used CRISPR to knock down Geminine. And by removing that brake, you unleash whatever residual CDT1 is left in the cell.
14:15Yes. That forces origin licensing to occur at entirely inappropriate times during the cell cycle. Wow. It completely reversed the resistance. And they didn't just test this with OLIPRIP. They validated it using AZD 5305.
14:28Oh, that's the highly selective next generation PRP1 inhibitor currently advancing through phase 3 clinical trials, right? Exactly. AZD 5305 is designed to have much lower hematological toxicity than older drugs because it selectively traps part P1 without affecting part 2.
14:44So when you remove Geminine, the cell's baseline replication stress skyrockets, the origins fire chaotically. And when you introduce the parp inhibitor on top of that drug induced chaos. The cancer cell is pushed far beyond its tolerance threshold.
14:59It attempts to segregate a massively fragmented, under replicated genome during mitosis. Which means the cell undergoes what we call mitata catastrophe and self destructs. Precisely. Using CRISPR to validate the pathway is elegant, but translating that to a human patient requires pharmacology.
15:15We need a small molecule that can perform that same brake cutting maneuver. And the researchers actually utilized a pharmacologic inhibitor called AF 615. This compound is specifically designed to disrupt the protein protein interaction between CDT1 and Geminine.
15:30So it chemically frees CDT one from sequestration. It does. Treating the highly resistant organoids with AF 615 in combination with new PARP inhibitor AZD 5305, fully restored drug sensitivity. That's huge.
15:44Yeah, even at vanishingly small doses, just one nanomeler of AZD 5305. The addition of the Gemini Inhibitor eradicated the resistance cells. The therapeutic logic is incredibly compelling. You screen a patient, identify the BRCA 2 mutation and the pre-RC loss, and instead of giving them a PRP inhibitor that you know will fail, you hit them with the PRP inhibitor plus AF 615 to forcibly revoke their resistance mechanism.
16:09It is a brilliant blueprint for rational combination therapy. However, the translational hurdles between this foundational biology and the standard of care are, well, they're substantial. Always are. What's the biggest roadblock?
16:21First, we need extensive longitudinal data from larger patient cohorts to definitively validate pre-RC copy number status as a predictive clinical biomarker for PRP inhibitor failure. Right. You'd need to retrospectively and prospectively track 1000s of patients to prove that this CDT1 or CDC 6 loss tightly correlates with progression free survival over years of treatment.
16:43Furthermore, we face a major infrastructural challenge in genomic medicine. Genes like CDT1, CDC 6, and DBF4 are not routinely included in standard commercial next generation sequencing panels for oncology.
16:55Wait, really? So if a patient's tumor is sequenced today at a major hospital, the oncologist won't even see the pre-RC copy number loss because the panel simply isn't designed to look for it. Exactly. We can't act on a biomarker if our diagnostic tools are blind to it.
17:10Yeah, that makes sense. The sequencing panels will have to evolve to incorporate these resistance genes before this combination therapy can be deployed effectively. The team also emphasized the need to determine if this mechanism is exclusive to the BRCA 2 deficient environment, or if tumors driven by BRCA1 mutations utilize the same pre-RC alteration to survive replication stress.
17:32Right, because the biology of BRCA1 and BRCE2 while related have distinct functional nuances and how they manage stalled forks. There is clearly more to uncover, but the central insight from this research fundamentally reframes our approach.
17:45Loss of pre-replication complex genes like CDT1 acts as an upfront shield against PRP inhibitors in BRCA2 mutant prostate cancer. By targeting the regulators of this complex, we might be able to strip away that shield and force the cancer to succumb to therapy.
18:00It emphasizes that resistance isn't always about the tumor actively mutating the drug's target. Sometimes, it's about the tumor passively altering its own cellular pacing to render the drugs mechanism of action irrelevant. By understanding that pacing, we can design therapies that force the cell into fatal metabolic errors.
18:20What does this mean for the future of personalized oncology? And could this resistance mechanism extend beyond prostate cancer to other therapies that induce replication stress? It's a question that could shape the next decade of cancer research.
18:34This episode was based on an open access article under the CCBY 4 license. You can find a direct link to the paper and the license in our episode description. If you enjoyed this, follow or subscribe in your podcast app and leave a 5 star rating.
18:49If you'd like to support our work, use the donation link in the description. Now stay with us for an original track created especially for this episode and inspired by the article you've just heard about.
18:58Thanks for listening and join us next time as we explore more science base by base.
