May 22, 2025 • 15 mins
Look in the mirror. What secrets about your health could your reflection be revealing? Far beyond counting wrinkles, scientists have developed an extraordinary AI system called FaceAge that transforms a simple photograph into a window to your true biological age – potentially revolutionizing how we understand health, disease, and longevity.
This groundbreaking technology has uncovered a startling reality: cancer patients appear almost five years older than their chronological age. More remarkably, this artificially-detected "older appearance" directly correlates with survival outcomes. For every decade older someone looks according to FaceAge, their mortality risk increases by 11-15%, even after accounting for cancer stage, treatments, and other clinical factors. What makes this finding particularly significant is that chronological age alone failed to predict these outcomes with similar accuracy.
Diving deeper into the biology, researchers discovered connections between facial aging patterns and variations in CDK6, a gene involved in cellular aging processes. This suggests FaceAge is capturing fundamental biological mechanisms rather than superficial appearances. Lifestyle impacts were equally revealing – current smokers appeared nearly three years older than former or never-smokers, providing visible evidence of how our choices accelerate biological aging. Perhaps most surprising was the finding that functional status (how physically capable someone is) didn't significantly correlate with facial age, suggesting FaceAge detects aging dimensions distinct from obvious physical decline.
What does this mean for your health journey? Understanding your personal biological aging profile could be the key to optimizing not just lifespan but healthspan – the quality years of life where you feel vibrant and well. This represents a paradigm shift from reactive healthcare to proactive longevity optimization. Ready to discover what your reflection might reveal about your biological age and health potential? Explore how these cutting-edge insights could transform your approach to wellness and longevity.
Disclaimer:
The information provided in this podcast is for educational purposes only and is not intended as medical advice. Always consult with a qualified healthcare professional before making changes to your supplement regimen or health routine. Individual needs and reactions vary, so it’s important to make informed decisions with the guidance of your physician.
Connect with Us:
If you enjoyed today’s episode, be sure to subscribe, leave us a review, and share it with someone who might benefit. For more insights and updates, visit our website at Lifewellmd.com.
Stay Informed, Stay Healthy:
Remember, informed choices lead to better health. Until next time, be well and take care of yourself.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
Here's something that might make you think People
with cancer tend to look well,on average, almost five years
older than they actually are.
It really makes you wonder.
You know, is that number onyour driver's license really the
best way to track your body'strue age, your vitality?
Speaker 2 (00:17):
That's a great point.
We're dealing with twodifferent ideas of age here,
aren't we?
Speaker 1 (00:22):
Right.
Speaker 2 (00:22):
You've got your chronological age, just how many
birthdays you've had, simpleenough.
Speaker 1 (00:26):
Yeah.
Speaker 2 (00:26):
But then there's biological age.
Speaker 1 (00:28):
Yeah.
Speaker 2 (00:28):
And that's well.
It's a much more nuancedpicture.
It reflects the actual state ofyour body, the wear and tear.
Yeah, you know, influenced byeverything your genes, how you
live, any sicknesses.
Speaker 1 (00:38):
And knowing that difference, the gap between
those two ages, that could bereally, really insightful,
couldn't it?
For understanding health, maybeeven predicting how someone
responds to treatment.
Speaker 2 (00:48):
Exactly so.
The big question then becomeshow do we actually get a good
measure, a reliable measure ofbiological age?
Speaker 1 (00:54):
Which brings us right to our deep dive today, I think
.
Speaker 2 (00:57):
It does.
We're looking at this reallycutting edge AI systems called
FaceAge FaceAge.
Speaker 1 (01:03):
Okay, and the idea which is pretty amazing is that
it looks at just a simplephotograph of your face.
Speaker 2 (01:09):
Just a photo.
Speaker 1 (01:09):
Just a photo and from that it estimates your
biological age.
There's a study on thisrecently.
Speaker 2 (01:15):
Wow, okay, so that sounds almost like something out
of a movie.
So our mission today, for youlistening, is to unpack this
FaceAge system.
We want to figure out how itactually sees a face.
You know what the science saysabout, how accurate it is and,
maybe most importantly, whatthis could all mean for us, for
understanding our own aging, ourown well-being.
Speaker 1 (01:37):
And this is especially relevant if you're
thinking about staying healthyand vibrant long-term.
Speaker 2 (01:42):
Right.
Speaker 1 (01:42):
At LifeWellMDcom here in florida.
That's precisely our focushelping people optimize not just
lifespan but health span.
You know the years you feelgood and function well that's
the key, isn't it?
Health span it is, andunderstanding biological aging
is absolutely fundamental tothat.
So innovations like face age.
Well, they could offer somereally valuable insights for
anyone interested in their ownwellness journey.
Speaker 2 (02:03):
Yeah.
Speaker 1 (02:04):
Definitely something to keep in mind as we go through
this.
So okay, let's get into it.
How does an AI look at apicture and guess, biologically
speaking, how old someone is?
What's going on under the hood?
Speaker 2 (02:16):
Well at its heart.
Face age is what we call a deeplearning system.
Speaker 1 (02:19):
Yeah.
Speaker 2 (02:20):
It basically learns from just vast amounts of data.
Think of it like teaching acomputer to spot patterns, but
on a huge scale.
In this case, it's been trainedby looking at tons of facial
features and photos and learninghow those features connect to a
person's age.
Speaker 1 (02:36):
And where did it get all these photos to learn from
Must have needed a lot.
Speaker 2 (02:39):
Oh yeah, Huge data sets.
It started with one called IMDBWiki that had photos of, I
think, over 58,000 people,mainly 60 and older, who were
presumed to be healthy.
Speaker 1 (02:50):
Okay, so a baseline of healthy aging faces.
Speaker 2 (02:52):
Right, and they used another set UTK Face for some
initial checks.
But what's interesting is howthey curated the data.
They really focused on the agerange you typically see in, say,
clinical oncology populations.
Speaker 1 (03:03):
So it's specifically tuned for that older demographic
relevant to cancer studies.
It's not just counting wrinkles.
Then what are the actual stepsthe AI takes?
Speaker 2 (03:13):
No, it's more complex than that.
There are essentially two mainstages.
First, it has to find the facein the photo accurately.
Speaker 1 (03:20):
Like zeroing in.
Speaker 2 (03:21):
Exactly.
It uses a specific method forthat, a kind of neural network,
and it's pretty good about 95%accurate in their tests at just
finding the face.
Speaker 1 (03:29):
Okay.
Speaker 2 (03:29):
Then step two a different, more advanced AI gets
involved.
This one analyzes the finerdetails textures, shapes,
shadows, all sorts of things,extracts the patterns linked to
age and uses those patterns tomake the age estimate.
Speaker 1 (03:43):
And how good was it, especially for that over 60
group you mentioned.
Speaker 2 (03:46):
In that clinically relevant age range.
Yeah, the average error wasabout 4.09 years.
Speaker 1 (03:51):
So plus or minus four years roughly.
Speaker 2 (03:53):
On average.
Yes, so if someone'scryologically 70, face age might
estimate between, say, 66 and74.
Yeah, which you know, for thiskind of technology is actually
quite impressive, it's on parwith, or maybe even better than,
some other models out there.
Speaker 1 (04:07):
Okay, estimating age from a face is one thing, but
the really fascinating part isthe link to health, particularly
cancer.
How did they connect face ageto actual cancer patient data?
Speaker 2 (04:19):
Right.
So the study took this face agetool and applied it to data
from several groups of cancerpatients, both in the
Netherlands and the US.
They had data from the MASTROstudy and also from cohorts at
Harvard one focused on thoraciccancers, another on palliative
care.
Speaker 1 (04:35):
And what did they do?
Compare the face age estimatesfor these patients to people
without cancer.
Speaker 2 (04:41):
Exactly, they compared the face age estimates
of the cancer patients to areference group of people
without cancer.
Exactly, they compared the faceage estimates of the cancer
patients to a reference group ofpeople without cancer.
Speaker 1 (04:47):
And what did they see ?
Did the cancer patients facessuggest they were biologically
older?
Speaker 2 (04:51):
They did.
Yes, yeah, quite clearly.
On average, the patients withcancer looked according to face
age, about 4.79 years older thantheir actual chronological age.
Speaker 1 (05:00):
Almost five years older Wow.
Speaker 2 (05:01):
Yeah, and that difference was statistically
significant, meaning it's veryunlikely to be just a random
fluctuation.
Speaker 1 (05:07):
So the takeaway is your face might actually be a
window into your underlyinghealth, especially when dealing
with something serious likecancer.
Speaker 2 (05:15):
It certainly seems that way, based on this research
.
Speaker 1 (05:18):
OK, now this is where it gets really powerful, I
think.
Did this looking older actuallypredict anything about how the
patients did their prognosis?
Speaker 2 (05:26):
It absolutely did.
They found a very clear linkLooking older as estimated by
face age was correlated withworse overall survival.
Speaker 1 (05:34):
Worse outcomes.
Speaker 2 (05:35):
Yes, and this held true across different cancer
types.
Studied pan cancer specificallyin the thoracic cancer group
and also in the palliative caregroup.
Speaker 1 (05:45):
How strong was that link?
Speaker 2 (05:50):
Well, they calculated that for every apparent decade
older someone looked accordingto face age, their risk of death
increased by about 11 to 15percent, depending on the group.
And, importantly, this wasafter they adjusted for other
known factors like cancer, stagetreatment etc.
Speaker 1 (06:01):
So it wasn't just that sicker people look older.
This face age score was addingindependent predictive
information.
Speaker 2 (06:07):
Precisely.
It showed significantindependent prognostic value
across different cancer typesand stages.
Speaker 1 (06:13):
That's huge.
Speaker 2 (06:15):
And what's potentially very impactful
clinically is how it might helpwith patients receiving
palliative care, those withincurable cancer.
Speaker 1 (06:23):
How so.
Speaker 2 (06:24):
Well, doctors already use models to estimate survival
in those situations right tohelp guide conversations about
care goals.
They found that adding the faceage estimate to those existing
models significantly improvedtheir accuracy.
The AUC, which is a measure ofpredictive accuracy, went up
quite a bit.
Speaker 1 (06:42):
So it could make those incredibly difficult
end-of-life predictions anddiscussions potentially more
accurate, more informed.
Speaker 2 (06:48):
That's the potential.
Yes, it suggests face age couldbe a really valuable
non-invasive tool to add to theclinical toolkit for those cuff
situations.
Improving the model's abilityto distinguish who might live
longer versus shorter iscritical there.
Speaker 1 (07:02):
Really powerful stuff .
And how did just using plainold chronological age stack up
in comparison?
Did that predict survival aswell?
Speaker 2 (07:10):
Interestingly, no, In many of the analyses the
patient's actual chronologicalage wasn't significantly linked
to survival once other factorswere considered.
And what's more, when theybuilt these predictive models,
adding face age generallyexplained more the variation in
survival outcomes than addingchronological age did.
Speaker 1 (07:29):
So biological age captured by the face seems more
relevant than just years lived,at least in this context.
Speaker 2 (07:35):
It really points in that direction.
Yeah, Face age seems to becapturing something deeper about
biological resilience ordecline.
Speaker 1 (07:41):
It's like our faces really are telling a story.
Now, you mentioned the studywent deeper, looking at the
molecular level too, trying tosee if face age connects to the
genetics of aging.
Speaker 2 (07:52):
Yes, that was another really interesting layer they
wanted to know is this facialappearance thing just
superficial or does it reflectsomething happening
fundamentally at the level ofour DNA and how it relates to
aging?
Speaker 1 (08:03):
How do they test that ?
Speaker 2 (08:05):
They analyzed DNA, specifically from lymphocytes,
which are immune cells, takenfrom patients in a Harvard
thoracic cohort.
These were patients withnon-small cell lung cancer.
Speaker 1 (08:14):
Okay.
Speaker 2 (08:15):
And they focused on genes already known to be
involved in cellular senescence,which is basically cellular
aging.
Speaker 1 (08:21):
Makes sense, start with the known players.
Speaker 2 (08:23):
Exactly.
They looked at a handful,initially like TERT ATM, p53,
genes involved in DNA repair andcell cycle control.
Then they used a tool calledgenomania to identify a wider
network of related genes.
They ended up analyzingvariations, called SNPs, in 22
different genes within thisaging network.
Speaker 1 (08:43):
Okay, so they're looking for links between
variations in theseaging-related genes and the face
age score.
What did they find?
Speaker 2 (08:49):
They found one significant hit After correcting
for testing multiple genes.
Face age showed a significantassociation with variations in a
gene called CDK6.
Speaker 1 (08:58):
CDK6.
And what about chronologicalage?
Did that link to any of thesegenes?
Speaker 2 (09:02):
Nope.
In their analysis,chronological age didn't show
any significant associationswith variations in these
specific aging pathway genes.
Speaker 1 (09:09):
Only face age, linked to CDK6.
What does CDK6 do?
Why is that potentiallyimportant?
Speaker 2 (09:14):
So CDK6 is a key regulator of the cell cycle.
It helps control when cellsdivide.
Speaker 1 (09:18):
Right.
Speaker 2 (09:18):
There's some evidence suggesting that CDK6 is a key
regulator of the cell cycle.
It helps control when cellsdivide Right.
There's some evidencesuggesting that CDK6 activity
might actually help delaycellular senescence or aging.
The finding here was an inverseassociation, meaning a higher
face age score was linked withgenetic variations potentially
associated with lower or alteredCDK6 function.
Speaker 1 (09:37):
Interesting.
So looking older might belinked genetically to a pathway
involved in controlling cellularaging speed.
Speaker 2 (09:44):
It's a tantalizing link.
Yeah, it hints that face agemight be tapping into some
pretty fundamental biologicalaging processes reflected in
gene activity.
Of course, they notedlimitations, like not having a
healthy control group for thisspecific gene analysis, but it's
definitely a strong lead forfuture research.
Speaker 1 (10:01):
Very cool.
Okay, so moving beyond the deepbiology, what about more
obvious factors?
We all know smoking can makeyou look older.
Did the study look at thingslike that?
Lifestyle factors influencingface age.
Speaker 2 (10:11):
They did.
Yeah, they looked at thedifference between face age and
actual age in one of the patientgroups, the master cohort, and
compared it across differentcharacteristics.
Speaker 1 (10:19):
And they confirmed.
Speaker 2 (10:22):
Yes, they reiterated that finding the face age
difference was significantlyhigher in cancer patients
compared to the presumed healthydata sets and other non-cancer
clinical groups they looked at.
Speaker 1 (10:33):
What about smoking specifically?
Speaker 2 (10:35):
That showed a really striking effect.
Current smokers lookedsignificantly older.
According to FaceAge, theaverage increase was over 33
months, so nearly three yearsolder.
Speaker 1 (10:46):
Wow, almost three years, just from current smoking
.
Speaker 2 (10:49):
Compared to former smokers and never smokers.
Yeah, it really highlights thevisible impact lifestyle choices
can have right there on yourface, reflecting potentially
accelerated biological aging.
Speaker 1 (10:59):
Any other factors?
What about weight, like BMI?
Speaker 2 (11:02):
They did find a statistically significant link
with BMI, but the effect sizewas actually quite small.
So while there's a connection,it wasn't nearly as pronounced
as something like smoking.
Speaker 1 (11:11):
Okay, and what about just general fitness or frailty?
Doctors use that ECOGperformance status score right.
Does being less functionallycapable line up with looking
older?
Speaker 2 (11:21):
according to FaceAge, that was another interesting
finding.
They looked at ECOG status,which ranges from fully active
to completely disabled, and theyfound no statistically
significant difference in theface age gap between the
different performance statusgroups.
Speaker 1 (11:35):
Really.
So someone could be quitedebilitated functionally, but
not necessarily look much oldervia face age or vice versa.
Speaker 2 (11:41):
It seems so, which suggests again that face age
might be capturing a dimensionof biological aging that's
distinct from functional statusor overall frailty as measured
by ECOG.
It's not just reflecting howsick someone appears in that
traditional sense.
Speaker 1 (11:57):
It's tapping into something else.
Speaker 2 (11:58):
Something potentially related more directly to
cellular aging processes,perhaps processes perhaps.
Speaker 1 (12:03):
So, pulling this all together, face age seems like it
has real potential to takesomething.
We notice subjectively how oldsomeone looks, and turn it into
objective, quantifiable data.
Speaker 2 (12:13):
Exactly.
Speaker 1 (12:14):
And that data seems to carry real clinical weight,
predicting survival in cancerpatients potentially better than
chronological age, and maybeeven linking back to molecular
aging pathways like CDK6.
Speaker 2 (12:24):
That's the promise.
Yes, it's a step towards usingeasily accessible information,
like a facial photo, to getdeeper insights into biological
age and its health implications.
It really underscores thispotential for understanding the
aging process better, maybeidentifying people at higher
risk for age-related issuessooner.
Speaker 1 (12:43):
It's translating what our eyes see into well
objective biological information.
Speaker 2 (12:48):
Precisely.
Of course, there's more workneeded.
Validation in bigger, morediverse groups is key, and we
need to think carefully aboutthe ethics.
Speaker 1 (12:56):
Ah, yes, like potential bias in the AI if it
wasn't trained on diverse enoughfaces or misuse by insurance
companies.
Speaker 2 (13:03):
Those are critical considerations.
The study did make efforts toassess and mitigate bias, but
it's an ongoing challenge.
With AI in healthcare, we needrobust safeguards and
transparency before somethinglike this becomes routine, and
correlating it further withother molecular aging markers
would strengthen the case too.
Speaker 1 (13:20):
Definitely important points, but the potential is
clearly there and for youlistening, you know this kind of
cutting edge research isexactly what excites us here at
LifeWellMDcom.
We're committed to exploringand utilizing the most advanced
science backed approaches tohealth, wellness and, especially
, longevity.
Understanding your personalbiological age is really
becoming a cornerstone of takingthat proactive, personalized
(13:43):
approach to staying well.
Speaker 2 (13:44):
Absolutely.
We believe that getting ahandle on your unique biological
aging profile empowers you.
It helps you make smarterchoices about lifestyle diet,
maybe even targetedinterventions to really optimize
your health span.
Speaker 1 (13:56):
Yeah.
Speaker 2 (13:57):
So if you're curious about your own health and
wellness journey and howunderstanding concepts like
biological age could fit in, wedefinitely encourage you to
reach out.
You can contact us atlifewellmdcom or give us a call
at 561-210-9999.
Speaker 1 (14:11):
Our team, guided by Dr Kumar, is ready to have that
personalized conversation andhelp you explore the latest in
health and longevity science.
Speaker 2 (14:17):
So let's wrap up with a final thought for everyone
listening.
Think about this how much doeswhat we see on the outside, our
appearance, truly mirror what'shappening deep inside,
biologically?
Speaker 1 (14:29):
And what incredible new insights might technologies
like face age unlock for usindividually and as a society,
as we all strive for healthier,longer lives?
It's definitely something toponder, isn't it?
Speaker 2 (14:41):
It really is.
The face might be telling usmore than we ever realized.
Here's something that might make you think People
with cancer tend to look well,on average, almost five years
older than they actually are.
It really makes you wonder.
You know, is that number onyour driver's license really the
best way to track your body'strue age, your vitality?
Speaker 2 (00:17):
That's a great point.
We're dealing with twodifferent ideas of age here,
aren't we?
Speaker 1 (00:22):
Right.
Speaker 2 (00:22):
You've got your chronological age, just how many
birthdays you've had, simpleenough.
Speaker 1 (00:26):
Yeah.
Speaker 2 (00:26):
But then there's biological age.
Speaker 1 (00:28):
Yeah.
Speaker 2 (00:28):
And that's well.
It's a much more nuancedpicture.
It reflects the actual state ofyour body, the wear and tear.
Yeah, you know, influenced byeverything your genes, how you
live, any sicknesses.
Speaker 1 (00:38):
And knowing that difference, the gap between
those two ages, that could bereally, really insightful,
couldn't it?
For understanding health, maybeeven predicting how someone
responds to treatment.
Speaker 2 (00:48):
Exactly so.
The big question then becomeshow do we actually get a good
measure, a reliable measure ofbiological age?
Speaker 1 (00:54):
Which brings us right to our deep dive today, I think
.
Speaker 2 (00:57):
It does.
We're looking at this reallycutting edge AI systems called
FaceAge FaceAge.
Speaker 1 (01:03):
Okay, and the idea which is pretty amazing is that
it looks at just a simplephotograph of your face.
Speaker 2 (01:09):
Just a photo.
Speaker 1 (01:09):
Just a photo and from that it estimates your
biological age.
There's a study on thisrecently.
Speaker 2 (01:15):
Wow, okay, so that sounds almost like something out
of a movie.
So our mission today, for youlistening, is to unpack this
FaceAge system.
We want to figure out how itactually sees a face.
You know what the science saysabout, how accurate it is and,
maybe most importantly, whatthis could all mean for us, for
understanding our own aging, ourown well-being.
Speaker 1 (01:37):
And this is especially relevant if you're
thinking about staying healthyand vibrant long-term.
Speaker 2 (01:42):
Right.
Speaker 1 (01:42):
At LifeWellMDcom here in florida.
That's precisely our focushelping people optimize not just
lifespan but health span.
You know the years you feelgood and function well that's
the key, isn't it?
Health span it is, andunderstanding biological aging
is absolutely fundamental tothat.
So innovations like face age.
Well, they could offer somereally valuable insights for
anyone interested in their ownwellness journey.
Speaker 2 (02:03):
Yeah.
Speaker 1 (02:04):
Definitely something to keep in mind as we go through
this.
So okay, let's get into it.
How does an AI look at apicture and guess, biologically
speaking, how old someone is?
What's going on under the hood?
Speaker 2 (02:16):
Well at its heart.
Face age is what we call a deeplearning system.
Speaker 1 (02:19):
Yeah.
Speaker 2 (02:20):
It basically learns from just vast amounts of data.
Think of it like teaching acomputer to spot patterns, but
on a huge scale.
In this case, it's been trainedby looking at tons of facial
features and photos and learninghow those features connect to a
person's age.
Speaker 1 (02:36):
And where did it get all these photos to learn from
Must have needed a lot.
Speaker 2 (02:39):
Oh yeah, Huge data sets.
It started with one called IMDBWiki that had photos of, I
think, over 58,000 people,mainly 60 and older, who were
presumed to be healthy.
Speaker 1 (02:50):
Okay, so a baseline of healthy aging faces.
Speaker 2 (02:52):
Right, and they used another set UTK Face for some
initial checks.
But what's interesting is howthey curated the data.
They really focused on the agerange you typically see in, say,
clinical oncology populations.
Speaker 1 (03:03):
So it's specifically tuned for that older demographic
relevant to cancer studies.
It's not just counting wrinkles.
Then what are the actual stepsthe AI takes?
Speaker 2 (03:13):
No, it's more complex than that.
There are essentially two mainstages.
First, it has to find the facein the photo accurately.
Speaker 1 (03:20):
Like zeroing in.
Speaker 2 (03:21):
Exactly.
It uses a specific method forthat, a kind of neural network,
and it's pretty good about 95%accurate in their tests at just
finding the face.
Speaker 1 (03:29):
Okay.
Speaker 2 (03:29):
Then step two a different, more advanced AI gets
involved.
This one analyzes the finerdetails textures, shapes,
shadows, all sorts of things,extracts the patterns linked to
age and uses those patterns tomake the age estimate.
Speaker 1 (03:43):
And how good was it, especially for that over 60
group you mentioned.
Speaker 2 (03:46):
In that clinically relevant age range.
Yeah, the average error wasabout 4.09 years.
Speaker 1 (03:51):
So plus or minus four years roughly.
Speaker 2 (03:53):
On average.
Yes, so if someone'scryologically 70, face age might
estimate between, say, 66 and74.
Yeah, which you know, for thiskind of technology is actually
quite impressive, it's on parwith, or maybe even better than,
some other models out there.
Speaker 1 (04:07):
Okay, estimating age from a face is one thing, but
the really fascinating part isthe link to health, particularly
cancer.
How did they connect face ageto actual cancer patient data?
Speaker 2 (04:19):
Right.
So the study took this face agetool and applied it to data
from several groups of cancerpatients, both in the
Netherlands and the US.
They had data from the MASTROstudy and also from cohorts at
Harvard one focused on thoraciccancers, another on palliative
care.
Speaker 1 (04:35):
And what did they do?
Compare the face age estimatesfor these patients to people
without cancer.
Speaker 2 (04:41):
Exactly, they compared the face age estimates
of the cancer patients to areference group of people
without cancer.
Exactly, they compared the faceage estimates of the cancer
patients to a reference group ofpeople without cancer.
Speaker 1 (04:47):
And what did they see ?
Did the cancer patients facessuggest they were biologically
older?
Speaker 2 (04:51):
They did.
Yes, yeah, quite clearly.
On average, the patients withcancer looked according to face
age, about 4.79 years older thantheir actual chronological age.
Speaker 1 (05:00):
Almost five years older Wow.
Speaker 2 (05:01):
Yeah, and that difference was statistically
significant, meaning it's veryunlikely to be just a random
fluctuation.
Speaker 1 (05:07):
So the takeaway is your face might actually be a
window into your underlyinghealth, especially when dealing
with something serious likecancer.
Speaker 2 (05:15):
It certainly seems that way, based on this research
.
Speaker 1 (05:18):
OK, now this is where it gets really powerful, I
think.
Did this looking older actuallypredict anything about how the
patients did their prognosis?
Speaker 2 (05:26):
It absolutely did.
They found a very clear linkLooking older as estimated by
face age was correlated withworse overall survival.
Speaker 1 (05:34):
Worse outcomes.
Speaker 2 (05:35):
Yes, and this held true across different cancer
types.
Studied pan cancer specificallyin the thoracic cancer group
and also in the palliative caregroup.
Speaker 1 (05:45):
How strong was that link?
Speaker 2 (05:50):
Well, they calculated that for every apparent decade
older someone looked accordingto face age, their risk of death
increased by about 11 to 15percent, depending on the group.
And, importantly, this wasafter they adjusted for other
known factors like cancer, stagetreatment etc.
Speaker 1 (06:01):
So it wasn't just that sicker people look older.
This face age score was addingindependent predictive
information.
Speaker 2 (06:07):
Precisely.
It showed significantindependent prognostic value
across different cancer typesand stages.
Speaker 1 (06:13):
That's huge.
Speaker 2 (06:15):
And what's potentially very impactful
clinically is how it might helpwith patients receiving
palliative care, those withincurable cancer.
Speaker 1 (06:23):
How so.
Speaker 2 (06:24):
Well, doctors already use models to estimate survival
in those situations right tohelp guide conversations about
care goals.
They found that adding the faceage estimate to those existing
models significantly improvedtheir accuracy.
The AUC, which is a measure ofpredictive accuracy, went up
quite a bit.
Speaker 1 (06:42):
So it could make those incredibly difficult
end-of-life predictions anddiscussions potentially more
accurate, more informed.
Speaker 2 (06:48):
That's the potential.
Yes, it suggests face age couldbe a really valuable
non-invasive tool to add to theclinical toolkit for those cuff
situations.
Improving the model's abilityto distinguish who might live
longer versus shorter iscritical there.
Speaker 1 (07:02):
Really powerful stuff .
And how did just using plainold chronological age stack up
in comparison?
Did that predict survival aswell?
Speaker 2 (07:10):
Interestingly, no, In many of the analyses the
patient's actual chronologicalage wasn't significantly linked
to survival once other factorswere considered.
And what's more, when theybuilt these predictive models,
adding face age generallyexplained more the variation in
survival outcomes than addingchronological age did.
Speaker 1 (07:29):
So biological age captured by the face seems more
relevant than just years lived,at least in this context.
Speaker 2 (07:35):
It really points in that direction.
Yeah, Face age seems to becapturing something deeper about
biological resilience ordecline.
Speaker 1 (07:41):
It's like our faces really are telling a story.
Now, you mentioned the studywent deeper, looking at the
molecular level too, trying tosee if face age connects to the
genetics of aging.
Speaker 2 (07:52):
Yes, that was another really interesting layer they
wanted to know is this facialappearance thing just
superficial or does it reflectsomething happening
fundamentally at the level ofour DNA and how it relates to
aging?
Speaker 1 (08:03):
How do they test that ?
Speaker 2 (08:05):
They analyzed DNA, specifically from lymphocytes,
which are immune cells, takenfrom patients in a Harvard
thoracic cohort.
These were patients withnon-small cell lung cancer.
Speaker 1 (08:14):
Okay.
Speaker 2 (08:15):
And they focused on genes already known to be
involved in cellular senescence,which is basically cellular
aging.
Speaker 1 (08:21):
Makes sense, start with the known players.
Speaker 2 (08:23):
Exactly.
They looked at a handful,initially like TERT ATM, p53,
genes involved in DNA repair andcell cycle control.
Then they used a tool calledgenomania to identify a wider
network of related genes.
They ended up analyzingvariations, called SNPs, in 22
different genes within thisaging network.
Speaker 1 (08:43):
Okay, so they're looking for links between
variations in theseaging-related genes and the face
age score.
What did they find?
Speaker 2 (08:49):
They found one significant hit After correcting
for testing multiple genes.
Face age showed a significantassociation with variations in a
gene called CDK6.
Speaker 1 (08:58):
CDK6.
And what about chronologicalage?
Did that link to any of thesegenes?
Speaker 2 (09:02):
Nope.
In their analysis,chronological age didn't show
any significant associationswith variations in these
specific aging pathway genes.
Speaker 1 (09:09):
Only face age, linked to CDK6.
What does CDK6 do?
Why is that potentiallyimportant?
Speaker 2 (09:14):
So CDK6 is a key regulator of the cell cycle.
It helps control when cellsdivide.
Speaker 1 (09:18):
Right.
Speaker 2 (09:18):
There's some evidence suggesting that CDK6 is a key
regulator of the cell cycle.
It helps control when cellsdivide Right.
There's some evidencesuggesting that CDK6 activity
might actually help delaycellular senescence or aging.
The finding here was an inverseassociation, meaning a higher
face age score was linked withgenetic variations potentially
associated with lower or alteredCDK6 function.
Speaker 1 (09:37):
Interesting.
So looking older might belinked genetically to a pathway
involved in controlling cellularaging speed.
Speaker 2 (09:44):
It's a tantalizing link.
Yeah, it hints that face agemight be tapping into some
pretty fundamental biologicalaging processes reflected in
gene activity.
Of course, they notedlimitations, like not having a
healthy control group for thisspecific gene analysis, but it's
definitely a strong lead forfuture research.
Speaker 1 (10:01):
Very cool.
Okay, so moving beyond the deepbiology, what about more
obvious factors?
We all know smoking can makeyou look older.
Did the study look at thingslike that?
Lifestyle factors influencingface age.
Speaker 2 (10:11):
They did.
Yeah, they looked at thedifference between face age and
actual age in one of the patientgroups, the master cohort, and
compared it across differentcharacteristics.
Speaker 1 (10:19):
And they confirmed.
Speaker 2 (10:22):
Yes, they reiterated that finding the face age
difference was significantlyhigher in cancer patients
compared to the presumed healthydata sets and other non-cancer
clinical groups they looked at.
Speaker 1 (10:33):
What about smoking specifically?
Speaker 2 (10:35):
That showed a really striking effect.
Current smokers lookedsignificantly older.
According to FaceAge, theaverage increase was over 33
months, so nearly three yearsolder.
Speaker 1 (10:46):
Wow, almost three years, just from current smoking
.
Speaker 2 (10:49):
Compared to former smokers and never smokers.
Yeah, it really highlights thevisible impact lifestyle choices
can have right there on yourface, reflecting potentially
accelerated biological aging.
Speaker 1 (10:59):
Any other factors?
What about weight, like BMI?
Speaker 2 (11:02):
They did find a statistically significant link
with BMI, but the effect sizewas actually quite small.
So while there's a connection,it wasn't nearly as pronounced
as something like smoking.
Speaker 1 (11:11):
Okay, and what about just general fitness or frailty?
Doctors use that ECOGperformance status score right.
Does being less functionallycapable line up with looking
older?
Speaker 2 (11:21):
according to FaceAge, that was another interesting
finding.
They looked at ECOG status,which ranges from fully active
to completely disabled, and theyfound no statistically
significant difference in theface age gap between the
different performance statusgroups.
Speaker 1 (11:35):
Really.
So someone could be quitedebilitated functionally, but
not necessarily look much oldervia face age or vice versa.
Speaker 2 (11:41):
It seems so, which suggests again that face age
might be capturing a dimensionof biological aging that's
distinct from functional statusor overall frailty as measured
by ECOG.
It's not just reflecting howsick someone appears in that
traditional sense.
Speaker 1 (11:57):
It's tapping into something else.
Speaker 2 (11:58):
Something potentially related more directly to
cellular aging processes,perhaps processes perhaps.
Speaker 1 (12:03):
So, pulling this all together, face age seems like it
has real potential to takesomething.
We notice subjectively how oldsomeone looks, and turn it into
objective, quantifiable data.
Speaker 2 (12:13):
Exactly.
Speaker 1 (12:14):
And that data seems to carry real clinical weight,
predicting survival in cancerpatients potentially better than
chronological age, and maybeeven linking back to molecular
aging pathways like CDK6.
Speaker 2 (12:24):
That's the promise.
Yes, it's a step towards usingeasily accessible information,
like a facial photo, to getdeeper insights into biological
age and its health implications.
It really underscores thispotential for understanding the
aging process better, maybeidentifying people at higher
risk for age-related issuessooner.
Speaker 1 (12:43):
It's translating what our eyes see into well
objective biological information.
Speaker 2 (12:48):
Precisely.
Of course, there's more workneeded.
Validation in bigger, morediverse groups is key, and we
need to think carefully aboutthe ethics.
Speaker 1 (12:56):
Ah, yes, like potential bias in the AI if it
wasn't trained on diverse enoughfaces or misuse by insurance
companies.
Speaker 2 (13:03):
Those are critical considerations.
The study did make efforts toassess and mitigate bias, but
it's an ongoing challenge.
With AI in healthcare, we needrobust safeguards and
transparency before somethinglike this becomes routine, and
correlating it further withother molecular aging markers
would strengthen the case too.
Speaker 1 (13:20):
Definitely important points, but the potential is
clearly there and for youlistening, you know this kind of
cutting edge research isexactly what excites us here at
LifeWellMDcom.
We're committed to exploringand utilizing the most advanced
science backed approaches tohealth, wellness and, especially
, longevity.
Understanding your personalbiological age is really
becoming a cornerstone of takingthat proactive, personalized
(13:43):
approach to staying well.
Speaker 2 (13:44):
Absolutely.
We believe that getting ahandle on your unique biological
aging profile empowers you.
It helps you make smarterchoices about lifestyle diet,
maybe even targetedinterventions to really optimize
your health span.
Speaker 1 (13:56):
Yeah.
Speaker 2 (13:57):
So if you're curious about your own health and
wellness journey and howunderstanding concepts like
biological age could fit in, wedefinitely encourage you to
reach out.
You can contact us atlifewellmdcom or give us a call
at 561-210-9999.
Speaker 1 (14:11):
Our team, guided by Dr Kumar, is ready to have that
personalized conversation andhelp you explore the latest in
health and longevity science.
Speaker 2 (14:17):
So let's wrap up with a final thought for everyone
listening.
Think about this how much doeswhat we see on the outside, our
appearance, truly mirror what'shappening deep inside,
biologically?
Speaker 1 (14:29):
And what incredible new insights might technologies
like face age unlock for usindividually and as a society,
as we all strive for healthier,longer lives?
It's definitely something toponder, isn't it?
Speaker 2 (14:41):
It really is.
The face might be telling usmore than we ever realized.
Popular Podcasts
United States of Kennedy
United States of Kennedy is a podcast about our cultural fascination with the Kennedy dynasty. Every week, hosts Lyra Smith and George Civeris go into one aspect of the Kennedy story.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com