Servier $210 million bet on IDEAYA's uveal melanoma therapy darovasertib seems to have paid off, as the drug has hit the mark in a phase 2/3 that could form the basis of regulatory filings later this year.
The results of the OptimUM-02 study showed that a combination regimen of PKC inhibitor darovasertib with Pfizer's cMET inhibitor Xalkori (crizotinib) achieved a statistically significant improvement in median progression-free survival (PFS) compared to a control group using investigators' choice of therapy, generally immunotherapies like MSD's Keytruda (pembrolizumab) and Bristol Myers Squibb's Opdivo (nivolumab) and Yervoy (ipilimumab).
The study enrolled previously untreated patients with HLA-A*A2:01-negative metastatic uveal melanoma (mUM), in other words, a group with the eye cancer that would not be eligible for treatment with Immunocore's Kimmtrak (tebentafusp), a TCR therapy that was approved by the FDA for HLA-A*02:01-positive mUM in 2022.
Shares in IDEAYA were up 16% on the trial results, which revealed that PFE with the darovasertib regimen was 6.9 months, versus 3.1 months in the control group. The objective response rate (ORR) came in at 37.1% and 5.8%, respectively, and there was also a trend towards improved overall survival (OS) that will continue to be monitored.
IDEAYA and Servier, which also pledged up to $320 million in regulatory and commercial milestones when it licensed rights to darovasertib last September, said that OptimUM-02 is the first randomised study to show a statistically significant and "clinically meaningful" PFS benefit in this patient population.
"Metastatic uveal melanoma is an area of high unmet medical need with poor prognosis and short overall survival, and there are currently no approved therapies for HLA-A*02:01-negative mUM patients," said Dr Meredith McKean of the Sarah Cannon Research Institute, who added that the results are "potentially practice changing."
According to IDEAYA, around 95% of patients with uveal melanoma have activating mutations in GNAQ/11 GTPase proteins that drive downstream PKC signalling and tumour growth. It estimates there are around 3,000 patients in the US, with approximately half of them progressing to mUM.
The company estimates that 50% to 70% of mUM patients have the HLA-A*02:01-negative serotype, suggesting that the market for darovasertib is at least as large as that for Kimmtrak, which brought in $400 million in sales last year.
Another treatment option is Delcath Systems' Hepzato (melphalan for injection/hepatic delivery system), which was approved by the FDA in 2023 specifically for uveal melanoma patients with unresectable liver metastases, which made around $80 million in 2025.
A Reuters report citing Truist analyst Gregory Renza said peak annual revenues for darovasertib could reach $800 million.
https://pharmaphorum.com/news/ideayaservier-pkc-drug-aces-uveal-melanoma-trial
The FDA is once again studying the safety of the abortion pill mifepristone just days after a federal Louisiana judge temporarily allowed the dispensing of the drug through mail.
“FDA continues to work on the collection of the robust and timely data that is necessary for a well-controlled study,” the agency wrote in an April 8 update to its website. The analysis will help the regulator decide whether it needs to make “substantive” changes to its Risk Evaluation and Mitigation Strategies for mifepristone, though it is unclear what such changes could look like.
“The agency is taking care to do this study properly and in the right way,” the FDA wrote, nevertheless adding that it wants to wrap up the study “as soon as possible.” Analyses such as these take “approximately a year or more” in academic settings, the regulator claimed, noting that “the current agency plan is to have this study done sooner than that.”
This new safety study comes a day after a Louisiana court blocked an attempt by Liz Murrill, Louisiana’s Attorney General, to prevent mifepristone from being dispensed to patients by mail. In a 37-page ruling on April 7, Judge David Joseph granted the FDA’s motion to press pause on the case while it conducts an assessment of mifepristone’s safety.
“The equities and the public interest weigh heavily in favor of FDA completing the job that the law requires it to do,” Joseph wrote. “At this juncture, it is the completion of FDA’s promised good faith, evidence-based, and expeditious review [of mifepristone’s safety] that this court finds to be in the public interest.”
Joseph, who was appointed by President Donald Trump, gave the FDA six months to provide the court “with the status of its review . . . and any updated timeframe for completion.”
Mifepristone is a progesterone blocker that is used to terminate pregnancies alongside another drug called misoprostol. It can also be used to treat high blood sugar in Cushing’s syndrome. Because of its role in inducing abortion, mifepristone has attracted controversy, particularly since federal protection of abortion rights was eliminated by the Dobbs v. Jackson Women’s Health Organization Supreme Court decision in 2022. In November 2022, for instance, the Alliance for Hippocratic Medicine filed a lawsuit against the FDA, asking to agency to pull its approval of the pill.
Then, in April 2023, Texas judge Matthew Kacsmaryk issued a preliminary injunction that blocked sales of mifepristone. This order kicked off a legal back-and-forth, at one point drawing an open letter from more than 480 biopharma leaders, and made it all the way to the Supreme Court. The letter called Kacsmaryk’s decision an “act of judicial interference” made “without regard for science or evidence.”
In June 2024, the High Court ruled to keep mifepristone available.
The fight over abortion access continued into 2025. In April, advocacy group Ethics and Public Policy Center published a report alleging that serious adverse events have developed in more than 10% of women on mifepristone. The report led health secretary Robert F. Kennedy, Jr. in May to tell the FDA to review its safety policies around the drug.
The anti-abortion report was “fundamentally, fatally flawed,” according to an open letter from 53 biotech executives and investors that asked the FDA to maintain access to mifepristone.
https://www.biospace.com/fda/fda-probes-abortion-pill-anew-after-court-keeps-mail-access-alive
by Steve Watson via modernity.news,
Residents in Atlanta are staring down robot dogs patrolling their own apartment parking lots — with live foreign operators apparently calling the shots through the machines.
Another viral video posted to X captures the raw street-level reaction as locals confront one of the mechanical units.
The operator responds in real time, and the accent leaves little doubt about the location of the person on the other end of the feed.
The post continues… “There are plenty of videos of these dogs calling the police on people, so that means someone sitting in India is patrolling our streets and calling the police on Americans. These robotic dogs are equipped with 360° cameras, thermal imaging, headlights, sirens, speakers, and sensors. Despite this they are not fully autonomous, they typically have a live human operator monitoring the feed remotely”
Another of these droids was seen recently giving commands to Americans in Atlanta. Even when citizens complied peacefully, the bot issued orders and summon real police — all while the eyes and ears behind the machines sit overseas.
People are seeing these things just roaming around like an episode of Black Mirror.
While in America they’re being used for parking lot security with the feed manned by foreigners somewhere, in China they’re taking the technology in a far more aggressive direction. Beijing has already unleashed machine-gun-toting robot wolves equipped with a “collective brain” for coordinated urban combat.
Separate footage shows an armed robot dog leading full combat drills as part of the global robot armies advance.
WATCH: Armed Robot Dog Leads Combat Drill; Global Robot Armies Advance
The contrast could not be clearer. Here at home, American neighborhoods get low-level security outsourced to foreign call-center workers who can now watch, listen, record, and report on U.S. citizens 24/7. In China, the same basic platform is being militarized for dominance on the battlefield.
These dogs are not toys. They carry 360-degree cameras, thermal imaging, and sensors that feed data straight back to whoever is paying the operator. That means sensitive footage of American homes, vehicles, and daily movements is streaming overseas — potentially stored, analyzed, or even shared with foreign governments. And when the bots decide to call the cops, it seems to be a foreign voice triggering American law enforcement against American citizens on American soil.
The push toward automation and remote monitoring is delivering a creeping surveillance state where the watchers aren’t even in the country. While elites celebrate the “future,” everyday Americans get robot dogs in the hood and foreign accents telling them to move along.
It’s only a matter of time before the novelty wears off and Americans start trashing these things, and then where do we go from there?
Just two days after a Molotov cocktail was hurled at OpenAI CEO Sam Altman's Russian Hill residence in San Francisco, another violent episode unfolded at the same address. Early Sunday morning, April 12, 2026, San Francisco police responded to reports of possible shots fired near Altman's home.
Around 3 a.m. on Sunday, April 12, officers responded to a report of possible gunfire in the 2000 block of the Russian Hill neighborhood. Surveillance footage and security reports indicated a Honda sedan drove past Altman's property; the passenger extended an arm from the window and fired one round toward the Lombard Street side of the residence. The vehicle fled, but its license plate was captured.
Investigators linked the car to Amanda Tom, 25 (the driver and registered owner), and Muhamad Tarik Hussein, 23 (the passenger). Both were arrested without incident on nearby Taylor Street. A search warrant executed at a residence yielded three firearms. The pair was booked into San Francisco County Jail on suspicion of negligent discharge of a firearm. No injuries were reported, and no one was inside the targeted area at the time of the shot.
SFPD has not publicly confirmed whether the shooting specifically targeted Altman or his home, describing it as an "apparent shooting near" the residence. Police Chief Derrick Lew emphasized the department's zero-tolerance stance: “The SFPD takes crimes involving guns extremely seriously and anyone committing acts like these will be arrested and prosecuted to the fullest extent of the law.”
Investigators pinpointed the owner of the car, identified as Amanda Tom, 25. She and Muhamad Tarik Hussein, 23, were arrested not far away, on Taylor Street, police said. Officers served a search warrant at a home and seized three guns, police said.
Tom and Hussein were booked into jail on suspicion of negligent discharge of a gun. -Chronicle
"The SFPD takes crimes involving guns extremely seriously and anyone committing acts like these will be arrested and prosecuted to the fullest extent of the law," said Police Chief Derrick Lew.
The shooting comes amid growing public anxiety over the societal impact of AI, from job displacement to the massive infrastructure demands of data centers. Last week's Molotov cocktail suspect drove to OpenAI headquarters and threatened to burn down the building.
On Friday at around 3:45 a.m., 20-year-old Daniel Alejandro Moreno-Gama allegedly threw an incendiary device at the metal gate of Altman's home on Chestnut Street. The device sparked a small fire that security quickly extinguished, with no injuries reported. Moreno-Gama then traveled to OpenAI's Mission Bay headquarters, where he allegedly threatened to burn down the building. He was arrested there using surveillance footage.
The suspect, originally from Texas, faced serious charges including attempted murder, arson, making criminal threats, and possession of destructive devices. He remains held without bail. Online writings attributed to him reveal deep concerns about AI as an existential risk: he described unaligned models as capable of deception and warned that tech leaders were "gambling with our future" without sufficient morals. He had engaged with groups like PauseAI but was not an active organizer calling for violence.
OpenAI confirmed the attack and expressed gratitude to the San Francisco Police Department (SFPD), noting heightened security measures. Altman himself responded hours later in a personal blog post, sharing a family photo with his husband and child. "I love them more than anything," he wrote, adding that he hoped the image might "dissuade the next person from throwing a Molotov cocktail at our house, no matter what they think about me." He acknowledged underestimating the power of "words and narratives," referenced a recent critical New Yorker profile amid "great anxiety about AI," and called for de-escalation: "While we have that debate, we should de-escalate the rhetoric and tactics and try to have fewer explosions in fewer homes, figuratively and literally."
As we noted on Friday, the timing and tone of Altman’s response appear to underscore a deeper reality now playing out across the country: financially strained American households are increasingly pushing back against the infrastructure demands of the AI industry. New data this week shows residential electricity prices surging in key regions, driven in large part by the explosive growth of data centers needed to train and run large language models.
Close enough https://t.co/reP3n5kJpR pic.twitter.com/PrH03ydD8A
— zerohedge (@zerohedge) April 10, 2026
Communities from Virginia to Georgia to the Midwest have mounted growing resistance - through zoning fights, moratoriums, and public hearings - over electricity costs, water consumption, land use, and limited local economic benefits, marking what one analysis described as a sharp escalation in Americans starting to revolt against data centers.
