Episode 3.03: The National Lung Matrix Trial [1:31]

• Background

  • Imatinib in 1999 was a revolutionary drug

    • It targeted BCR-ABL, which was the key oncogenic event in CML and the sole driver
      • It is crazy that the first Tyrosine kinase inhibitor ended up being the best Tyrosine kinase inhibitor

  • Imatinib plus the Human Genome Project allowed us to sequence the genome cancer cells in widespread fashion formulated an explosion of targeted drugs

    "When you put the two things together, an explosion of targeted drugs propelled by the first and luckily best druggable target, and the ability to sequence tissue, you had the perfect storm for precision oncology" - Dr. Prasad

• Estimation of the percentage of us patients with cancer who benefit from genome-driven oncology

  • Paper by Marquart et al. in JAMA Oncology

    

    Marquart et al.

• **The National Lung Matrix Trial of personalized therapy in lung cancer [12:50]**

  • Published in Nature by Middleton et al.

• Bayesian adaptive design

  • This Bayesian analysis has three go/no go endpoints:
    1. Disease control rate
    2. Response rate
    3. PFS
  • The sample is not representative in a few ways:
    1. The researchers have to exhaust proven treatment options
    2. The researchers are also selecting for the kinds of people who decline some of these therapeutic options
    3. The researchers are filtering for people who have a unique molecular alteration
  • For these reasons, these are worthless endpoints in an uncontrolled study where there are too many selection filters inserted

• A Pharmacologic and Toxicological Study of Amygdalin

  • Paper by Charles Moertel
    • This paper dismantled the idea that amygdalin would benefit cancer patients
  • Response rate means activity, it means a drug has promise that you could consider it for trial of efficacy
    • Response rate was developed because we needed a way to prioritize what to test
      • If a drug has no response it's a "no go," like amygdalin
        • Drugs that lack single-agent activity: are they worth pursuing in combination?

• Applicability

  • Out of 5,467 patients with samples sent to SMP2 → 302 patients with actionable targets allocated to a trial drug

• Response Rate

  

  Twitter Thread

• Takeaway

  • This is useful information but to use this information to create new drugs for common mutations is a flawed takeaway message
    • The conclusion is not "we need better drugs," it is "is this a useful strategy?"
      • The majority of these arms failed, which means that drugging mutations in very complex tumors is really unlikely to significantly bend the curve

Episode 3.06: ADAURA [22:38]

• Osimertinib in Resected EGFR-Mutated Non–Small-Cell Lung Cancer
  • After surgical therapy, patients are randomly assigned in a 1:1 ratio to receive either osimertinib or placebo for 3 years.
    • For another look at this study–refer to Episode 2.55
• Staging
  • If CT Scans are being conducted for stage 2 & 3 patients–this is below the US SoC
    • "Brain imaging was performed" is too vague

      • CTs can not be used to stage–an MRI of the brain is needed to stage

        "If you don't do adequate staging, to look for metastatic disease, you very likely are including some occult metastatic disease" - Dr. Prasad

• 3A Lung Cancer
  • These patients often go to interdisciplinary tumor boards and post-surgery, they often receive some consolidative radiotherapy
    • It may be the case that this trial has a prohibition on that
      • This would further increase relapse rates needlessly in the control arm
• The chemotherapy rate
  • We don't want a trial of osimertinib versus placebo in people who are being tricked to not get chemotherapy
    • Offering the trial could easily dissuade someone from chemotherapy–which is an intervention that has a durable DFS benefit
• What drugs are given upon metastatic presentation?

  • Dr. Prasad has a sneaking suspicion that this trial is going to be giving people who present in the control arm with metastatic disease gefitinib or lorlatinib

    "If your control arm is essentially malpractice, then the trial was not a useful study." - Dr. Prasad

Episode 3.06: IMbrave150 [1:26:10]

• Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma
  • This trial randomly assigned patients in a 2:1 ratio to receive either atezolizumab + bevacizumab vs. sorafenib
• Background
  • Hepatocellular carcinoma has very poor outcomes, especially when it is unresectable
    • Currently, two category one agents are approved in the frontline setting for unresectable hepatocellular carcinoma
      1. Sorafenib in Advanced Hepatocellular Carcinoma
         • Median overall survival was 10.7 months in the sorafenib group and 7.9 months in the placebo group
      2. Lenvatinib
         • Based on the REFLECT non-inferiority trial
• Population
  • Patients had to have had no prior systemic therapy and measurable disease as defined by risk 1.1
    • biopsy was not mandated in this study
    • All of the patients were had preserved liver function (Child–Pugh class A)
    • ECOG of 0-1