AI summary by TwinMind
Mikkael Sekeres | When Blood Breaks Down: Life Lessons from Leukemia | Talks at Google
Talks at Google
55m 12s
Introduction and Background
- Dr. Mikhail Sekris: Director of leukemia program at Cleveland Clinic, professor of medicine, vice chair for clinical research at Taussig Cancer Institute
- Author of "When Blood Breaks Down: Life Lessons from Leukemia"
- Writes regularly for New York Times Well section
- Discussion topics: book, doctor-patient communication, cancer treatment changes during COVID-19
Writing Motivation and Medical Storytelling
- Family background: father was journalist at Providence Journal (Rhode Island)
- Medicine based on storytelling: patients tell illness stories to doctors/nurses who pass them to colleagues/trainees
- Leukemia as "malignant golem" - terrible invader corrupting body's normal machinery
- Inspiration from patients making impossible treatment choices with grace and dignity
- Book chronicles real-life doctor-patient journeys through leukemia diagnosis
Leukemia Mechanics and Biology
- Cancer = uncontrolled cell growth from genetic abnormalities (inherited or acquired)
- Mutations give cells growth/survival advantage, prolonged lifespan
- Bone marrow leukemia: cells fill narrow bone space ("high rent district"), normal cells die
- Normal bone marrow produces: red blood cells (oxygen transport), white blood cells (infection fighting), platelets (bleeding control)
- Paradox: sky-high white count (leukemia cells) but low red cells/platelets (normal cell death)
- Acute leukemia: cells stuck in "eternal childhood" - immature, non-functional despite high counts
Book Structure and Patient Composites
- Three main patient stories arriving within one day
- Joan: 48-year-old surgical nurse, mother from Wooster, Ohio
- David: 68-year-old factory worker, husband/father
- Sarah: 36-year-old pregnant woman with leukemia diagnosis
- All composites of real patients over nearly two decades
- Every conversation, medical twist, life aspect actually occurred
- Started with "point of tension" - gut feeling something doesn't feel right
- David's tension: older adult choosing aggressive chemotherapy (4-6 weeks hospitalization, <10% cure chance) vs. less intensive therapy (0% cure, <5% death risk)
- Sarah's tension: balancing treatment for herself vs. fetal effects
- Joan's tension: treating rare acute leukemia with emerging Chinese treatments in real-time
Treatment Discoveries and Research History
- Curative therapy from Chinese root discovery
- Dog/beagle experiments leading to bone marrow transplantation
- Brian Druker's research: focused on one drug for rare disease, lost job, eventually led to greatest leukemia therapy discovery
AI and Machine Learning in Medicine
- Working with Aziz Nazi on AI approaches for leukemia
- Pathologist agreement problem: only 75% agreement on bone marrow diagnoses
- Local vs. expert pathologist disagreement: 25-40% of time
- Machine learning applications:
- Bone marrow snapshot analysis for diagnosis
- Prognostic schemes incorporating genetic mutations (95% of leukemia patients have identifiable mutations)
- Predicting drug response based on genetic mutation profiles
- Moving from 1950s chromosome analysis (physical photo cutting) to sophisticated genetic mutation identification
Electronic Medical Records and Technology
- "Workstations on wheels" (formerly "computers on wheels") - renamed to avoid patient offense
- Mixed blessing of electronic medical records:
- Positive: reduces chemotherapy errors through pre-built order sets with calculated doses
- Negative: doctors staring at screens instead of patients, "pajama time" interference
- Challenge: cut-and-paste culture eliminates thoughtful writing/processing of patient information
- Research difficulties: Epic built as billing system, not research system, requires data dumping to "Clarity"
- Number one physician dissatisfaction factor: electronic medical records and time spent in them
Risk Communication and Patient Decision-Making
- Emergency room scenario: patient thinks they have flu, gets leukemia diagnosis, told "get affairs in order"
- Complex information processing: explaining leukemia biology + treatment options within 24 hours
- Treatment choice example: <20% cure chance with high death risk vs. 0% cure with <5% death risk
- Emotional/family pressure: spouse and children influencing decisions
- Staged information delivery: present information, let process, return with more
- Patient coping mechanism: 75% remember best odds when asked about cure chances (said >50% when told <10%)
- Bioethics question: true informed consent when patients misremember/reframe statistics?
Neurodivergent Patient Communication
- Case: ~50-year-old man with intellectual capacity of 6-8 year old (oxygen deprivation at birth)
- Jehovah's Witness beliefs: no blood products or risk hell
- Mother as power of attorney helping decisions
- Bioethics consultation involvement
- Challenge: ensuring patient understanding of dying vs. preserving hope of someone mentally 6-8 years old
- No clear answers in bioethics for these situations
COVID-19 Impact on Cancer Care
- Dramatic drop in new cancer center patients during peak COVID
- Fear: delayed diagnoses leading to more advanced cancer stages
- Misdiagnosis case: leukemia patient with flu-like symptoms diagnosed as COVID, placed in quarantine, leukemia diagnosis delayed
- Cancer patients with COVID: many times more likely to die (especially leukemia patients due to immune system dysfunction)
- Advanced cancer patients choosing hospice over ICU when getting COVID
- Healthcare settings implementing extreme safety precautions
- Message: don't delay screening or symptom evaluation
Treatment Decision-Making and Specialization
- Super narrow specialist focusing only on leukemia/related bone marrow cancers
- Responsibility: know emerging treatments, participate in clinical trials, anticipate FDA approvals
- Rarely surprised by new therapies due to specialization
- Treatment timeline planning: therapy A now, therapy B in couple years, therapy C later
- Incorporate new breakthrough treatments as soon as appropriate
Patient Reading: Mrs. Abrams Story
- First day of internship in ICU with resident Dave
- Mrs. Abrams: early 40s pediatrician with metastatic ovarian cancer
- Chose not to be intubated, decided to die breathing on own
- Children ages 10 and 8 said goodbye
- Final night: writing birthday, holiday, graduation cards for next 10 years
- Cards for every occasion so she could remain part of children's lives
- Died writing cards all night with husband and friend helping
- Pivotal moment leading to oncology career choice
- Connection to people facing terrible cancers with dignity
Supporting Cancer Patients
- Don't treat patients "like glass" - maintain normal interactions and teasing
- Address elephant in room: ask about cancer or preference not to discuss
- Avoid generic "anything you need, just ask"
- Offer concrete help: specific dinner delivery days, transportation to appointments
- Be deliberate about communication preferences and boundaries
- Some want to discuss cancer constantly, others want safe zones without cancer talk
Screening and Prevention
- Cancer screening complexity: designed to "overcatch" and over-diagnose
- False positive likelihood higher than finding unknown cancer in asymptomatic people
- US approach: discrete screening points vs. regular screening
- Breast cancer example: 50% lifetime chance of false positive requiring further testing
- Politics involved: organizations with vested interests advocate younger screening ages
- US screening guidelines generally good but payment/access issues
- Investment problem: screening benefits realized 10-20 years later, beyond current job tenure
COVID Vaccine and Leukemia Patients
- Vaccine timeline prediction: second quarter 2021 availability, September 2021 herd immunity
- "Needles in arms" challenge: development, manufacturing, distribution, convincing people
- Most COVID vaccines are dead viruses (safe for leukemia patients) vs. live attenuated (dangerous)
- Unknown: whether leukemia patients will develop immunity from vaccines
- Plan: give vaccines when approved but maintain precautions due to uncertain effectiveness
Leukemia Cure Statistics and Outcomes
- Adult acute leukemia cure rate: 30-40% (all comers)
- Pediatric acute lymphoblastic leukemia: >90% cure rate (greatest cancer therapy success story)
- Treatment protocol: 4-6 weeks hospitalization, then outpatient treatment
- 6 months intensive chemotherapy, potentially 1.5-2 years lower dose maintenance
- Cure definition: alive without leukemia 5 years post-diagnosis
- Cured patients have essentially normal lifespans
- Older adults face "competing risks" from other health conditions (heart disease, diabetes)
- Once leukemia gone, it's gone permanently
