Ming-Hei Tai, PharmD, BCOP’s Post

TROPION-Lung01 was recently published in JCO.  Patients with metastatic non-small cell lung cancer (NSCLC) who progressed on platinum-based chemoimmunotherapy were randomized to datopotamab deruxtecan (dato-DXd) 6 mg/kg or docetaxel 75 mg/m2 once every 3 weeks.  Median progression-free survival (PFS) was statistically superior (4.4 vs 3.7 months, HR 0.75 [0.62-0.91]) but overall survival (OS) was not (12.9 vs 11.8 months, HR 0.94 [0.78-1.14]).  Docetaxel appears to remain unbeaten in the second line chemotherapy setting - a 0.7 month "improvement" in PFS with no OS benefit isn't something that will change clinical practice. In the subgroup analysis, investigators found that in nonsquamous NSCLC, PFS was 5.5 vs 3.6 months (HR 0.63 [0.51-0.79]) and OS was 14.6 vs 12.3 months (HR 0.84 [0.68-1.05]).  So, a numerically better HR for PFS and OS when compared to the overall population, but with OS still not being statistically significant.  It's also important to note that studies aren't powered for subgroups. This trial reminds me of the pemetrexed vs gemcitabine trial (https://lnkd.in/edwfaM7q) published in JCO back in 2008.  In that noninferiority trial, pemetrexed was found to be noninferior to gemcitabine in NSCLC.  A subgroup analysis found pemetrexed have a statistically superior OS in adenocarcinoma, with p = 0.03, barely making the cutoff.  That trial is why pemetrexed is a preferred agent in adenocarcinoma, despite that subgroup not being properly powered, and no phase III trial in adenocarcinoma being performed to confirm the subgroup results. As this was a superiority trial, and there was no statistically significant OS benefit in any subgroup, it is unlikely that dato-DXd will be approved.  An interesting question is, what if this was a noninferiority trial like the pemetrexed trial?  Could this have been an option in second-line nonsquamous NSCLC, similar to how pemetrexed squeezed its way into nonsquamous NSCLC?  Safety wise, treatment-related adverse events (TRAEs) were 87.5% vs 86.9%, and 25.6% vs 42.1% for grade ≥3 TRAEs.  Docetaxel had more dose reductions and discontinuations, but dato-DXd had more treatment interruptions.  The main reason why docetaxel had more grade ≥3 TRAEs was neutropenia.  So if both agents were viewed as "equivalent" efficacy wise, it's possible that it could have been viewed as a less toxic agent for some patients. Overall, what can we conclude?  Docetaxel is not to be underestimated, and has consistently shown efficacy in multiple phase III trials.  Dato-DXd works in lung cancer, but its role in this space remains unclear.  A fully powered trial in nonsquamous NSCLC should be performed. #lungcancer https://lnkd.in/evPAVM3Y

Some studies just sit there waiting to be closed.  For example, NRG-GI004 is technically open for enrollment at Karmanos, and looks at giving atezolizumab in dMMR colorectal cancer.  However, nobody has enrolled on this trial recently, and it is probably similar for the other 444 locations where NRG-GI004 is open.  According to clinicaltrials.gov, which was last updated August 28th, the trial is still recruiting, with a target of 120 patients.  This trial has tough competition - several agents are commercially available, and the PD-1 inhibitor based trials are much more attractive.  The window for clinician willingness to enroll patients on this trial has probably passed.  One of the reasons why clinical trials in the oncology space are so expensive is because a lot of money is spent to avoid logistical concerns and speed up enrollment.  For example, some studies will send study drug to every site before any patient enrolls, just to simplify the ordering process.  Not every site will use all the drug that is sent to them, meaning a portion of the supply that is sent out is wasted.  It's more expensive to lose a trial due to low enrollment than it is to write off unused medications, but that adds to the cost of a clinical trial.  Every clinical site probably has a story about the biggest clinical trial write off they saw. #clinicaltrials #oncology #tecentriq https://lnkd.in/eNUAnMTQ

Did you know that 18 pathologists only had 26% concordance on whether a HER2 IHC stain was 0 or 1+?  This study, done back in 2022, collected 170 biopsies from Yale and sent the scanned slides to 18 different pathologists.  Of those 170 biopsies, 92 were read as IHC 0 by at least one of the 18 pathologists.  17 of 18 pathologists agreed on 24/92 of these cases (26%), meaning that you have a good chance of turning an IHC 0 patient into an IHC 1 patient simply by sending the biopsy slides to another pathologist.  That would qualify a breast cancer patient for fam-trastuzumab deruxtecan - no repeat biopsy needed! Differences in HER2 testing before and after fixation have also been noted in our national guidelines (https://lnkd.in/eA8qKshR).  This is why central HER2 testing is very important in studies that look at HER2-directed agents, especially in "low" and "ultralow" patients, because it eliminates variation between institutions. New HER2 testing methodologies are needed (and being studied), but then the question is, how would that impact fam-trastuzumab deruxtecan?  The current FDA approval may need to be re-validated for any new testing methodology, not to mention all the clinical trials that currently use HER2 IHC testing.  #breastcancer #enhertu https://lnkd.in/eJh5XeCA

Very interesting perioperative lung data from Neocoast-2, which was a 5-arm phase II trial.  Three of the arms were presented at IASLC.   The first arm was neoadjuvant durvalumab (D) + chemotherapy (CT) + oleclumab (O), a novel anti-CD73 checkpoint inhibitor, followed by adjuvant D + O. The second arm was neoadjuvant D + CT + monalizumab (M), a novel anti-NKG2A checkpoint inhibitor, followed by adjuvant D + M.  The third arm was neoadjuvant D + CT + Dato-DXd, a TROP-2 directed antibody-drug conjugate (ADC), followed by adjuvant D.  Patients on the third arm received a single platinum agent, as opposed to doublet platinum chemotherapy. According to the abstract, pCR rates were 20%, 26.7%, and 34.1% respectively in the three arms presented.  Assuming that neoadjuvant chemoimmunotherapy has around a 20-25% pCR rate, the two double checkpoint inhibitions arms appear to show no improvement over currently available options.  These are small sample sizes, so double checkpoint inhibition with novel agents could still have a benefit, but that would need to be examined in a phase III trial.  In addition, the apparent benefit from the D + CT + Dato-DXd arm is from a higher pCR rate in the PD-L1 TPS <50% and the PD-L1 negative subgroups,  However, there are also weird subgroup analyses - for example, in the D + CT + O arm, 5.6% of patients with a PD-L1 TPS of <50% had a pCR, but 17.6% of PD-L1 negative patients had a pCR. It's hard to make any conclusions from these subgroups. The side effect profile appeared to be manageable.  Grade 3 or higher treatment-related adverse events were 33.8%, 38.0% and 18.5% respectively.  8.1%, 12.7%, and 7.4% of patients discontinued neoadjuvant therapy, so most patients appear to have finished all planned courses of treatment.  Overall, what can we conclude from this?  D + CT + Dato-DXd showed promising efficacy, but it's important to remember that this arm only had 44 patients.  One of the issues with reading efficacy from any 5-arm study like this is that they have wide confidence intervals, so the actual pCR rate could be much lower or higher than what was reported.  Dato-DXd could be a preoperative agent, but that will need to be compared against a standard-of-care chemoimmunotherapy regimen in a phase III trial.  #imfinzi #opdivo #keytruda #lungcancer https://lnkd.in/eSp3MXRQ

This phase II trial of atezolizumab before and after chemoradiation, published in JAMA oncology, is an interesting take on the idea of combining immunotherapy with radiation.  In this trial, patients received atezolizumab, then radiation, then consolidation chemotherapy, then atezolizumab maintenance. The disease control rate at 12 weeks was 74.2%. Median progression-free survival was 30.0 months and median overall survival was not reached. How does this trial relate to current trials? Recently, combination immunotherapy plus radiation was found to not be better than radiation followed by immunotherapy in the PACIFIC-2 trial (https://lnkd.in/eGSsw5hQ). However, that trial did not look at immunotherapy before chemoradiation, only in combination. In terms of trials that looked at immunotherapy before radiation, the trials I found were related to ipilimumab in melanoma (https://lnkd.in/eJw7Rzke and https://lnkd.in/eCTPvUYW). There could be other trials out there, but at the very least, there are no large trials that have tested this hypothesis. One concern is delaying treatment, but we are already using neoadjuvant therapy and delaying treatment for surgically resectable lung cancer patients. At some point, hopefully we have a large trial that at least tests this hypothesis of immunotherapy before radiation. #tecentriq #lungcancer https://lnkd.in/e5YQcZbQ

We don't often see phase III trials in leiomyosarcomas, due to its rarity.  The NIH doesn't even have an estimate of the incidence (https://lnkd.in/ey2rrWCf), although other sources estimate that this occurs in about 1 in 100,000 people. This phase III trial, published in the NEJM, enrolled 150 patients with metastatic or unresectable leiomyosarcoma who had not received prior chemotherapy.  Patients were randomized to either doxorubicin or doxorubicin plus trabectedin followed by maintenance trabectedin.  The study previously only reported progression-free survival (PFS), but now can report overall survival (OS) as well.  For overall survival, doxorubicin/trabectedin had an adjusted hazard ratio of 0.65 [0.44-0.95], 33 months vs 24 months.  37% of patients on the doxorubicin arm received second line trabectedin, and an additional 22% received trabectedin third line or later, so crossover seems to be appropriate.  100 patients received some sort of subsequent therapy, and for these patients, time to second progression was significantly longer in the doxorubicin/trabectedin group (HR 0.46 [0.32 to 0.65]), but this does include patients who did not get second line trabectedin.  Overall, either in terms of first or second progression or overall survival, patients on the experimental arm appear to have done better. What does this mean for sarcomas?  It's good because we finally have a new chemotherapy regimen with an overall survival benefit in leiomyosarcoma.  However, the time it took to enroll 150 patients (enrollment started Jan 2017) highlights the challenges in this space.  Sarcomas are both rare and heterogenous, so most trials have to be run in specific subtypes.  This is one of those tumors where the initial referral should absolutely be to an expert at an academic medical center.  Patients need to get enrolled onto clinical trials, and there's isn't an app you can click on which tells you what clinical trials are available, and which ones the patient qualifies for.  Otherwise, we are mostly relying on treatments which may not have changed for decades.  #yondelis #sarcoma https://lnkd.in/e_vz5kFh

This data on minimal residual disease (MRD) in first line treatment of chronic lymphocytic leukemia, published in JCO, shows that minimal residual disease (MRD) probably shouldn't be a surrogate endpoint in CLL.  End-of-treatment MRD had a weak correlation with overall survival, and this makes sense, considering that many CLL patients do well even if they are MRD positive.   Furthermore, looking at the supplemental data, peripheral blood MRD had a much poorer correlation with OS, meaning that if this was adopted as a surrogate endpoint, patients would need to undergo bone marrow biopsies to determine their MRD, which isn't common in clinical practice.   And ultimately, the problem here is that MRD is mostly used in time-limited therapy.  We still don't know how a time-limited MRD guided therapy would compare to an indefinite duration therapy. #cll https://lnkd.in/gGm4yshB

One question remaining, now that we are using perioperative immunotherapy in "hot" tumors, is whether we can skip surgery completely for some patients.  For example, in SWOG S1801, 21% of melanoma patients who received perioperative pembrolizumab had a complete pathological response.  In addition, only 6% of patients had a complete imaging-based response, showing that imaging-based response is not very specific.  Is there a combination of modalities that would allow us to better identify patients who have a complete pathological response, or at least bring us closer?  That would allow us to figure out whether these patients could skip surgery.  Right now, the published phase III clinical trials we have make all patients proceed to surgery irrespective of any biomarkers.  Surgery has its own complications, and it would be great if we could identify some method of finding patients who do not benefit from surgery after immunotherapy.  #immunotherapy #melanoma #keytruda https://lnkd.in/gJmBH6T5

Data on postrecurrence therapy from CheckMate-238 was published in the JCO. This was a trial that looked at giving adjuvant nivolumab or ipilimumab in resected melanoma. Data like this helps us understand when we might be able to rechallenge a patient after treatment with immunotherapy. Looking at the data, I would note these are small sample sizes. For example, only 11 patients who progressed on nivolumab received combination PD1/CTLA4 inhibition. There is also a selection bias, because providers may be more likely to give dual checkpoint inhibition to patients who are more able to tolerate it, and those patients may be predisposed to better outcomes. So what second line treatment should we be choosing? It seems clear that we should be giving BRAF/MEK inhibition to patients who have a BRAF mutation. As for dual vs single checkpoint inhibition, the data is unclear. It appears as though dual checkpoint inhibition did slightly better than single checkpoint inhibition, but that is confounded by selection bias and sample size. So it mostly supports current clinical practice of giving dual checkpoint inhibition to patients who can tolerate it. In terms of time to recurrence, it appears that patients who had recurrence within 12 months had a poorer response to therapy. So these may be patients who we would want to find clinical trials for, including TIL therapy. Patients who had recurrence after 12 months appear to have done well, and can be treated with currently available options. In summary, this study doesn't change clinical practice, but suggests that we should look at clinical trials for patients who progress within 12 months. #opdivo #yervoy #melanoma https://lnkd.in/edd6rSzF

We are probably reaching the end of how far we can push immunotherapy in lung cancer.  KEYNOTE-867 was recently discontinued by Merck.  There was a lack of event-free survival (EFS) benefit at interim analysis, per the press release.  This was a trial that looked at adding pembrolizumab to stereotactic body radiotherapy (SBRT) in patients with stage I or II non-small cell lung cancer (NSCLC).  The typical therapy for these patients would be surgical resection, so this trial only included patients with inoperable tumors, patients who were too ill to receive surgery, or patients who refused surgery. According to the press release, pembrolizumab was associated with higher rates of adverse events (AEs), including AEs leading to death, when compared to placebo.  The subgroup analyses may provide a lot of data for clinicians, since this was a fairly unique patient population.  It will be interesting to see if there were any differences between subgroups depending on disease stage, ECOG status, and reason for not receiving surgical resection.  The increase in adverse events should also remind us that immunotherapy isn't a harmless treatment, even though it is less toxic than chemotherapy.  Patients can develop long term complications like diabetes or hypothyroidism from immunotherapy, as well as acute toxicities that lead to death. #pembrolizumab #lungcancer https://lnkd.in/egM6RhND