Subscribe
Share
Share
Embed
Truth Seekers: Where Data Meets Reality
Tired of sensational headlines and conflicting health advice? Join Alex Barrett and Bill Morrison as they cut through the noise to uncover what scientific research actually says about the claims flooding your social media feed.
Each week, Alex and Bill tackle a different health, nutrition, or wellness claim that everyone's talking about. From "blue light ruins your sleep" to "seed oils are toxic," they dig into the actual studies, examine the methodologies, and translate the data into plain English.
No agenda. No sponsors to please. No credentials to fake. Just two people committed to finding out what's really true by going straight to the source—the research itself.
Perfect for anyone who's skeptical of influencer health advice but doesn't have time to read every scientific study themselves. New episodes drop regularly, delivering clarity in a world full of clickbait.
Question everything. Verify with data. Find the truth.
Disclaimer: Truth Seekers provides educational content based on published research. Nothing in this podcast should be considered medical, financial, or professional advice. Always consult qualified professionals for decisions affecting your health and wellbeing.
**When Your COVID Vaccine Becomes Cancer Treatment: Too Good to Be True?**
Alex: Alright, so cancer patients getting immunotherapy are twice as likely to be alive three years later if they got a COVID vaccine. That's the headline going around right now.
Bill: Yeah, and this isn't some fringe website either. This is Nature, MD Anderson Cancer Center—
Alex: Right, which is exactly why people are taking it seriously. The headline makes it sound like we've accidentally stumbled onto a cancer breakthrough while trying to fight COVID.
Bill: The numbers look incredible. Lung cancer patients, melanoma patients, all showing dramatically better survival if they happened to get an mRNA COVID vaccine around the time they started immunotherapy.
Alex: But here's what I'm wondering—when I was working in media, any time a headline sounded this definitively good, there was usually a gap between what the study showed and what the coverage claimed.
Bill: Oh, there's a gap here.
Alex: Brilliant. Let's find it.
Bill: So first, what they actually did. Researchers at MD Anderson looked back through their records from 2019 to 2023. They found 180 lung cancer patients who got COVID vaccines within 100 days of starting immunotherapy, and compared them to 704 patients who didn't get vaccinated.
Alex: Okay, looked back through records. So this is retrospective?
Bill: Exactly. They're looking at what already happened, not randomly assigning some patients to get vaccines and others not to.
Alex: And that distinction matters because...
Bill: Because when you look backwards, you're comparing two groups of people who were already different from each other in ways you might not be measuring.
Alex: Hang on—what do you mean already different?
Bill: Think about who got COVID vaccines during this period. These patients had to be alive, healthy enough to receive a vaccine, motivated enough to get one, and have access to vaccination. That's not a random group.
Alex: So you're saying the vaccinated patients might have been healthier to begin with?
Bill: Or more engaged with their healthcare. Or had better access to medical care. Or were more adherent to treatment protocols. All of those things predict better cancer outcomes completely independent of whether the vaccine does anything.
Alex: Wait, this sounds familiar. Didn't we—hang on, didn't we cover something like this? The vaccine and dementia thing?
Bill: Oh yeah. The Shingrix study.
Alex: That's the one. Same problem, wasn't it? People who got vaccinated were just more health-conscious to start with?
Bill: Exactly the same pattern. Selection bias masquerading as treatment effect.
Alex: Huh. Okay, so that's... that's quite a confound then.
Bill: And it gets more complicated. The study controlled for 39 different variables—things like age, cancer stage, smoking status, all that.
Alex: Which sounds impressive.
Bill: It is! The statistics are sophisticated. But here's the thing—they can only control for variables they measured. There are probably dozens of unmeasured factors that differ between someone who chose to get vaccinated during cancer treatment and someone who didn't.
Alex: So when we see that hazard ratio of 0.51, meaning vaccinated patients had roughly half the death risk—
Bill: —we're seeing a real statistical association in this dataset. But association isn't causation.
Alex: Right, but everyone says that. What's the actual alternative explanation here?
Bill: Okay, check this out. Let's say vaccinated patients were just more health-conscious or had better healthcare access. They might be more likely to attend follow-up appointments, take medications as prescribed, report side effects early, maintain better nutrition—all things that improve cancer outcomes.
Alex: So the vaccine isn't doing anything; it's just a marker for "person who engages well with healthcare."
Bill: That's one possibility the retrospective design can't rule out.
Alex: But they must have thought about this, right? MD Anderson isn't publishing in Nature without considering selection bias.
Bill: They absolutely considered it. In fact, here's what's really interesting—the researchers themselves explicitly say this is hypothesis-generating research that needs to be validated in a Phase III randomized trial.
Alex: Wait, the researchers are saying "we need an actual trial to prove this"?
Bill: Yes. Direct quote from their press release: "These findings have prompted a randomized Phase III trial to determine if mRNA COVID vaccines should be part of the standard of care."
Alex: So they're not claiming this proves anything yet.
Bill: Not at all. They're saying, "We found an interesting association, here's a plausible mechanism, now let's actually test it properly."
Alex: But that's not what the headlines say.
Bill: No.
Alex: The headlines say "turbo-charges cancer treatment" like it's established fact.
Bill: And that's the translation gap. The researchers are being appropriately cautious; the media coverage is not.
Alex: Okay, so what would an actual randomized trial look like?
Bill: You'd prospectively enroll cancer patients starting immunotherapy and randomly assign them to either get an mRNA vaccine or a placebo. Neither the patient nor the oncologist would know who got what. Then you follow them for years and see if survival differs.
Alex: And why can't we just trust the retrospective data?
Bill: Because randomization eliminates selection bias. If you flip a coin to decide who gets the vaccine, then both groups should be identical on average—not just on the 39 variables you measured, but on everything. Health-consciousness, socioeconomic status, treatment adherence, all of it.
Alex: Right.
Bill: When I was doing A/B testing at the tech company, this was the whole point. You randomize because you can't measure every possible confounding variable. The coin flip does the work for you.
Alex: So randomization solves the "who chose to get vaccinated" problem.
Bill: Exactly. But we don't have that data yet. We have an interesting observation that might reflect a real biological effect, or might reflect who chose to get vaccinated.
Alex: Okay, let me play devil's advocate, though. They also did mouse studies showing the mechanism works. Doesn't that support the human findings?
Bill: It does support biological plausibility. They showed that mRNA vaccines trigger interferon responses that can enhance checkpoint inhibitors in mice with melanoma and lung cancer.
Alex: So the mechanism makes sense.
Bill: The mechanism is scientifically sound. Type I interferons can activate immune cells and reprogram the tumor microenvironment. That part is real biology.
Alex: But...
Bill: But mice in a lab aren't cancer patients in hospitals.
Alex: Right, but hang on—if we can see the mechanism working in mice, doesn't that at least make the human association more credible? Like, it's not just random noise in the data if there's actual biology backing it up.
Bill: Okay, I'd push back on that a bit.
Alex: Yeah?
Bill: Mouse immune systems are different. Mouse tumors are different. And most cancer treatments that work brilliantly in mice fail when tested in humans.
Alex: Most?
Bill: The vast majority. Something like 95% of drugs that show promise in animal models fail in human clinical trials.
Alex: That's... actually that's quite bad.
Bill: Preclinical studies are important for understanding mechanism, but they're terrible predictors of clinical efficacy.
Alex: Mmm. Okay, but I guess what I'm saying is—doesn't it move this from "pure selection bias" to "plausible hypothesis worth testing"?
Bill: Oh, absolutely. I'm not saying the mouse studies are worthless. They show the mechanism could work. But they don't validate the retrospective human data. We still need the randomized trial.
Alex: Right, okay. That's fair.
Bill: Actually, let me back up. I think I'm being too dismissive of the mouse work. You're right that it makes the hypothesis more plausible. It just doesn't solve the selection bias problem in the human study.
Alex: So we have retrospective human data that might be confounded, and mouse data that might not translate, but together they're enough to justify a proper trial.
Bill: Yeah, that's better. That's a fair summary.
Alex: Right, but here's what I'm actually wondering—why does this matter? If cancer patients see this headline and ask their oncologist about getting a COVID vaccine, what's the harm?
Bill: That's a great question. COVID vaccines are safe for most cancer patients, and they obviously protect against COVID, which is particularly dangerous for immunocompromised people.
Alex: So getting vaccinated makes sense anyway.
Bill: Absolutely. But there's a difference between "you should get a COVID vaccine for COVID protection" and "you should get a COVID vaccine to treat your cancer."
Alex: Because one is proven and one isn't.
Bill: Right. And if patients start thinking of COVID vaccines as cancer treatment, they might make decisions based on that—delaying proven therapies, expecting specific anti-cancer effects that may not materialize, or feeling like they've failed if they can't access vaccines.
Alex: Or feeling devastated if they got the vaccine and their cancer still progressed.
Bill: Exactly. The emotional stakes are high when you're dealing with cancer. False hope is genuinely harmful.
Alex: Mmm. So what should cancer patients actually take from this research?
Bill: I think there are a few key points. First, this is interesting early-stage research showing a potential association. It's not proof of causation, and it's not ready for clinical implementation.
Alex: Second?
Bill: The biological mechanism is plausible enough that a randomized trial is justified. If you're interested in this, watch for those trial results in the next few years.
Alex: And third?
Bill: Get your COVID vaccines for COVID protection, especially if you're immunocompromised. That's established medicine. Just don't expect them to treat your cancer unless and until we have randomized trial data showing they do.
Alex: What's frustrating is that the research itself seems quite good. The problem is the gap between what it shows and how it's being presented.
Bill: That's what makes this a perfect example of science communication failure. The researchers did solid work within the limitations of retrospective analysis. They're calling for proper trials. But somewhere between the Nature paper and the headlines, "interesting association that needs testing" became "COVID vaccines cure cancer."
Alex: And that transformation happens in the headline writing, not in the research.
Bill: Every time. The incentive structure in media rewards definitive-sounding claims, not accurate representations of evidence hierarchies.
Alex: Which is why people end up confused about what science actually shows.
Bill: Yeah. And it's particularly problematic in cancer, where people are desperate for good news and vulnerable to overpromising.
Alex: Actually, that reminds me—wait, did I say 180 lung cancer patients earlier? Is that right?
Bill: For the lung cancer cohort, yeah. But they also looked at melanoma patients. Different numbers there.
Alex: Right, okay. I just wanted to make sure I wasn't misremembering the study design.
Bill: No, you got it. 180 vaccinated lung cancer patients compared to 704 unvaccinated.
Alex: Anyway, what were we saying about the incentive structures?
Bill: Oh, just that the media rewards definitive claims. Which creates this gap between careful research and breathless headlines.
Alex: So the takeaway isn't "this research is garbage"—
Bill: No, the research is valuable. It's hypothesis-generating. It might lead to a real breakthrough if the randomized trials confirm the association.
Alex: The takeaway is "this research shows an interesting pattern that needs proper testing before we call it a treatment."
Bill: And that distinction—between an observation and a proven treatment—is exactly what gets lost in translation.
Alex: For listeners dealing with cancer or supporting someone who is, I think the key message is: don't make treatment decisions based on retrospective studies. Wait for the randomized trials. Ask your oncologist about established treatments. And get your COVID vaccines for the reason they're proven to work—protecting against COVID.
Bill: And watch this space. If those Phase III trials show the association holds up under randomized conditions, that would genuinely be a breakthrough.
Alex: But until then, we're at the "interesting finding" stage, not the "proven treatment" stage.
Bill: Which is exactly where the researchers themselves say we are. We should probably listen to them.