Devraj Basu, MD, PhD, FACS

@JNCI_Now We defined the gene expression profiles distinguishing HPV+ OPSCCs that are prone to recur after TORS-based therapy. These the tumor-intrinsic and immune-related traits were tightly interrelated and generalizeable to nonsurgical cases as well. academic.oup.com/jnci/advance-a…

New @Nature study: >50% of lung cancer metastases are seeded by other metastases, not the primary tumor. This "seeding from seeding" reveals a complex evolutionary cascade that allows cancer to colonize the body. nature.com/articles/s4158…

It is a well-written editorial, and I agree with some parts and disagree with others. However, I am very disappointed that @TheLancet would choose a member of the scientific advisory board of IBA (a proton manufacturer) to write an editorial about protons. The author of an important commentary should not have such an intrinsic conflict of interest with the subject matter.

@weldeiry I have as yet to see an an AI tool that can mimic reasoning using basic biologic principles. Imagine if you were a student today trying to use it to understand those principles.

I asked Google AI “is it a good idea to use unmutated oncogenic HPV E6 as an antigen to vaccinate against HPV?” The answer “Using unmutated, wild-type oncogenic HPV E6 as a vaccine antigen is generally considered unsafe for human clinical use because of its potential to transform cells. While the E6 oncoprotein is a potent target for immunotherapy, it must be de-oncogenized (mutated) to eliminate its ability to bind and degrade p53 while retaining immunogenicity to be effective and safe.”

Now published: New study on the controversial role of confirmatory HPV testing in p16+ OPCs 📚 Corresponding author: @BasuLab1 Read more: ascopubs.org/doi/10.1200/PO… link in our bio 🔗 #AcademicMedicine #Research #HPV #headandneckcancer

@LocasaleLab A misrepresentation of lO drug activity - for a few cancers, they are profoundly life prolonging and occasionally curative for otherwise incurable disease. For many other cancer types, the profound effects are limited to maybe 5pct, with another 10-30pct deriving some benefit.

Pharma and academic scientists are correct. These immunotherapy drugs are real breakthroughs. When they work, they can cure patients with otherwise terminal cancer. The reality, unfortunately, is that they only work in about 1–5% of cancer patients. Yet there is relentless hype that these drugs are the endgame for cancer. Much of the pharma and biotech industry has oriented itself around them, and a large share of NIH funding follows. Meanwhile, 95–99% of cancer patients, for whom these drugs do nothing, are left behind. That is the part people are not paying attention to.

📢Thoughtful discussion of the utility of #protontherapy "[...] importance of integrating patient-reported outcomes into trial design and focusing on endpoints that matter most to long-term survivors. The goal is [...] thoughtful selection of patients" thelancet.com/journals/lance…

Come for the critically important data on protons for oropharyngeal cancer, stay for the beautifully written paper. TORPEdO is out, and it is fascinating and instructive. I will warn you upfront that this is a long thread, but there is a lot here to discuss! thelancet.com/journals/lance…

Background: We cure most locally advanced OPSCC patients with IMRT + cisplatin, but late dysphagia, xerostomia, dysgeusia, and weight loss still wreck QoL. Protons spare OARs beautifully on paper. Does that actually help patients? TORPEdO was built to answer exactly that.

Our new study on the controversial role of confirmatory HPV testing in p16+ OPCs found 6% RNAScope false negatives but no true negatives, supporting only selective, CAP guideline-based use of confirmatory tests when pre-test probability of HPV-relatedness is high. ascopubs.org/doi/10.1200/PO…

@Jdcramer has made a critical point below about KEYNOTE-689. The EFS rules in KN-689 require careful review to interpret these results.  A protocol-specific event was determined by blind independent central review (BICR), but if growth in the pre-surgical CT was perceived to be a “flare” (i.e. potential progression), surgery was supposed to proceed unless unresectable (protocol quote below). In order to be considered an event, repeat imaging was required 4-8 weeks later, and obviously if the tumor was still resectable the patient would go to surgery before then. This definition means it was extremely difficult to have an event before surgery, even if the tumor grew and the surgery was more extensive than initially anticipated. Thus some patients may have been harmed by neoadjuvant treatment, but we would not see that in the event data. Thus there is this important disconnect: 82 patients in the pembro arm stopped the drug due to progression (by the investigator, as shown in the CONSORT diagram), even though there were only 69 progression events (by BICR, used for endpoint analyses). Is perioperative immunotherapy doing something favorable and important in a subset of these patients? Yes, and hopefully we can refine its use to the right population and with the right regimen (adding neoadjuvant chemo or RT?). Is it distinctly possible/probable that some patients are progressing on neoadjuvant pembro, leading to a worse outcome in some domain, but we cannot see that in these data? Yes, unfortunately, also true.

What REALLY makes a good oncologist? in @JCO_ASCO After 10 years and 203 JCO essays, one thing became clear: expertise alone isn’t enough. Patient-centered care. Clear, authentic communication. Emotional intelligence. And the harder work Showing up again and again. Openness to uncertainty. Carrying the burden of cancer care. Continuous growth. Clinical competence is expected Human connection makes the difference. Science treats disease. Presence treats fear. ascopubs.org/doi/pdf/10.120… @OncoAlert

@VpwndF @BenMazer 🤣

@BenMazer FALSE. Marty just told me last week they want me back full-time at UCSF immediately so that I could continue my important trust-building operations

My most unconventional view is that I think people who wrote stuff like this hold a sliver of blame for the loss of public trust.

Federal funding for US biomedical research is moribund. Since October 1 2025, NIH is -80% in new grants and -70% in values (total dollars). Labs are closing down and researchers are leaving science. To what end?

At long last, the surgical outcomes from KN671! Some very useful and informative data within this manuscript: annalsthoracicsurgery.org/article/S0003-…

It was a privilege to present today at the 9th annual DDR Summit in Boston , representing the @PennCancer team advancing translational science and clinical trials for HPV+ head and neck cancer @DiabLabPenn @DrNickLukens @LovaSunMD

Two papers in @NatureCancer highlighting the role of intratumoral bacteria & associated immune architecture (e.g., neutrophil influx) in mediating resistance to immunotherapy in head & neck cancer. nature.com/articles/s4301… nature.com/articles/s4301… @GeneCollector @xrtGenomics

The field is understandably desperate to improve outcomes in this high-risk population of HPV-negative HNSCC. However, we also have to be cautious in interpreting phase II studies. This sub-100 patient trial led to a truly immense amount of additional work (and money) that ultimately led nowhere. Sometimes, that will happen, despite the best intentions (as in this case). A few final thoughts: 1. Choose endpoints wisely, and make them meaningful for clinical inference and the follow-up trial. It is atypical to use alive with locoregional control (with a chi-squared test) as a primary endpoint, especially for a systemic drug with a variety of other activities (good or bad). 2. Power the study with sufficient numbers to ask the relevant question. It was quite small, with baseline imbalances in a critical clinical characteristics, with unfortunate, random deaths further compromising interpretation. Smaller trials are cheaper and faster (and the preference for efficiency is understandable), but they can lead to this exact scenario. 3. When transitioning from a phase II to III, don’t change the experimental arm unless there are real data supporting it. I doubt the addition of the 3 extra cycles did anything – the study was bound to be negative – but there was no reason to do it, and patients may have been hurt if it increased the metastatic potential of the tumor. 4. Ensure there are safety guardrails. The interim analysis was performed after 65% of total events. Many trials use 50%, and that would have saved time (and maybe patient enrollment) here. Given the small phase II study, the group should have looked at the results sooner, before enrolling 700+ patients. 5. I deeply share the impatience with wanting to improve outcomes, but more caution is needed in interpreting phase II outcomes, especially when there are intrinsic limitations to the study itself. 6. It is an imperative to deploy more creative phase II (and phase III) designs to more quickly and accurately identify novel, successful therapies.

The optimal management of oligometastatic HNSCC is a debated and important question. How can we improve *overall survival* in this more favorable population? EA3211 is a phase III randomized trial studying this exact issue. Enrollment is even more streamlined with the most recently activated amendment: - Patients may be enrolled 3 months after completing the first course of radiotherapy - Patients may be enrolled after starting a compliant chemoimmunotherapy regimen (chemo dose prior to enrollment at discretion of physician) - Timelines for imaging more flexible given logistical challenges of work-up Please contact me with any questions! We are always looking to expand enrollment!

K-means is one of the most powerful algorithms for data scientists. But it's confusing for beginners. Let's fix that: