Hannah R. Abrams
LightNews LightNews
hannahrabrams.bsky.social
Hannah R. Abrams
@hannahrabrams.bsky.social
2.3K followers 380 following 150 posts
Heme/Onc fellow at Fred Hutch, hoping to reduce costs & improve patient/caregiver experiences in cancer care. Social Media Editor @ JCO OP. Views are my own. #MedSky
Posts Replies Media Videos
hannahrabrams.bsky.social
Important aspect of recent lung cancer screening discussion prompted by the new JAMA Open article: even if we were to keep pack-year criteria, current cutoffs are inadequate & disproportionately rule-out Black patients.

From 2019 analysis: ~30% sensitivity, 80% specificity
Figure 1. Distribution and Median Smoking Pack-Years at Diagnosis by Race for Incident Lung Cancer Cases. Both figures show higher incidence of lung cancer in lower pack-year distribution for African American people compared to White people. Sensitivity and specificity of 20 versus 30 pack year cutoffs for lung cancer screening guidelines. For African American people, sensitivity at age 55 with 30 pack year cutoff is approx 30%, specificity 80%.
November 25, 2025 at 8:39 PM
hannahrabrams.bsky.social
Wow. Data from @n8pennell.bsky.social & team #JCOOP confirming the care fragmentation many of us see in real life.

Patients with metastatic NSCLC whose docs felt it was reasonable for them to restart treatment after rehabbing at a SNF:

- Only 54% saw outpt onc again
- 31% restarted tx
Highlighted passage: Included in this study were patients with biopsy-confirmed mNSCLC who had not yet progressed or experienced serious adverse events to single-agent chemotherapy, and thus had reasonable options for further palliative systemic therapy. Patients must have been discharged to a SNF with an intention for further therapy once functional status permitted. Fig 1. Sankey diagram depicting planned systemic therapy, proportion of patients who received said therapy, and treatment response. Chemo, any chemotherapy containing regimen; ICI, immune checkpoint inhibitor; TKI, tyrosine kinase inhibitor. Of the total 427 patients, 173 (40.5%) had a hospital readmission within the subsequent 30 days after discharge to a SNF. Only 229 (53.6%) ever managed to return to an outpatient oncology visit, and only 131 (30.7%) received any subsequent systemic therapy. This proportion varied by the intended therapy (Fig 1; Table 2), with 40 of the 54 planned for a TKI (74.1%), 42 of the 130 planned for ICI (32.3%), and 49 of the 243 planned for cytotoxic chemotherapy (20.2%) receiving further therapy (P < .001).
November 21, 2025 at 4:50 AM
hannahrabrams.bsky.social
In Sept, Bishal Gyawali published in #JCOOP on the high cost of some 'more convenient' Rx formulations (peg-GCSF, oral GNRH instead of SQ/IM, combo pills)

Out now, European perspective on how to operationalize this:

ascopubs.org/doi/10.1200/...
November 19, 2025 at 3:49 AM
hannahrabrams.bsky.social
Exciting end to #CTOS2025 with final presentation from Erica Pimenta. Could PPARγ-2 or GLP1 agonists restore differentiation in dedifferentiated liposarcoma?
Scientist at lectern with slide reading "Q&A" Conclusion slide: • Intact IGF1/PPARG2 in WDLPS • PPARG2 restores differentiation in DDLPS GLP1 signaling is the dominant, druggable epigenetic program in DDLPS Induces increased expression of adipose markers Identifying chromatin modifiers at the PPARG2 TSS Fully characterizing GLP1-induced differentiation and other cellular processes modified through this axis
November 15, 2025 at 9:43 PM
hannahrabrams.bsky.social
Can adding aromatase inhibitor (+/-LHRHa) rescue patients with #PEComa who progress after responding to mTOR inhibitor? #CTOS2025

- n=23 women
- 26% PR, 35% SD
- Time to progression generally shorter than initial mTORi response
- Some exceptions in patients with TSC1/TSC2/PTEN
November 14, 2025 at 4:17 PM
hannahrabrams.bsky.social
.@RobinL_Jones @royalmarsden presenting ChonDRAgon PhII registrational study of Ozekibart in conventional #chondrosarcoma #CTOS2025.

- No approved systemic therapies
- mPFS 5.5 mo, with 21.7% w/o prog @ 12 mo
- Activity in both IDH-WT & mut
- TRAE incl ~13% fatigue, ~6% hepatitis
Man at lectern smiling ChonDRAgon study design. Patients allowed to crossover to ozekibart at progression KM curve with PFS at 12 mo of 21.7% Tornado plot of TRAE with fatigue, diarrhea, nausea, and elevated LFTs highest
November 14, 2025 at 2:45 PM
hannahrabrams.bsky.social
#CTOS2025 Need options for pts w/ angiosarcoma who are refractory to ICI. In PhI, injected oncolytic virus w/ TGF-B trap (AdAPT-001) shows some efficacy! Notably heavily pre-tx group.
List of sites and prior treatment lines for patients enrolled. All patients had received both chemotherapy and checkpoint inhibition. Receding scalp lesion shown in photographs Multiple receding lesions shown.
November 13, 2025 at 8:01 PM
hannahrabrams.bsky.social
More data on TCRs in synovial sarcoma/MRCLS! In pooled analysis, Lete-cel (NY-ESO-1), w/ similar ORR to afami-cel (MAGEA4) & median OS 20.5 mo w/ small subset long-term responders.

Exciting to have multiple cell therapy options, need further studies to know how to seq & cross-sensitivity. #CTOS25
Waterfall plot of response rate with lete-cell in SS/MRCLS with 43% ORR Paired waterfall plot and KM curves of afami-cel and lete-cel
November 13, 2025 at 7:25 PM
hannahrabrams.bsky.social
Larotrectinib for TRK-fusion sarcomas: when it works, it works.

- 38% long-term (2 year) responders
- In responders, 73% remain without progression at 6 years, 98% OS

#CTOS2025 Next questions: When to stop? How to predict responders?
Waterfall plot of change in max lesion size among patients with TRK fusion sarcomas who had a 2+ year response
November 13, 2025 at 3:56 PM
hannahrabrams.bsky.social
Wisdom from #JCO assoc editor Bob Maki at #CTOS25:

“There are many ways to lose your shirt. Las Vegas, Crypto, and [bad] statistical design.”
A man stands at a podium in front of a slide, the top of which reads "Conclusions: there are many ways to lose your shirt"
November 12, 2025 at 6:58 PM
hannahrabrams.bsky.social
Where to go from here?

You'll have to read the whole article in #JCO/#JCOOP, but here's a sneak preview!

Overall, supporting new ICIs & cross-label Rx may help ⬇️ price, and I think it's good there's incentive to explore new indications (incl rare cancers) w newer ICIs.
ascopubs.org/doi/10.1200/...
Additionally, the government could address inadequate competition by funding comparative effectiveness studies for different drugs in the same class or mandating standard- of-care active comparators. Generating evidence that drugs could be prescribed cross-label (ie, one drug being pre- scribed for an indication for which a different within-class drug is approved) would enable more direct competition within indications.
November 12, 2025 at 12:52 AM
hannahrabrams.bsky.social
Metastatic 1st line (EGFR/ALK-WT) #NSCLC has by far the most ICIs approved, with 7.

Notably, the later approvals did add either unique combinations (#1 durva/treme + platinum backbone by histology; #2 atezo + carbo/taxol +/- bev)

ascopubs.org/doi/10.1200/...
November 12, 2025 at 12:52 AM
hannahrabrams.bsky.social
Early comers, like pembro & nivo, also hold more unique approvals. As new drugs have come onto the market, many have been studied in areas with 1-2 existing indications.

ascopubs.org/doi/10.1200/...
Table showing number of total indications and unique (non-overlapping) indications for ICIs by mechanism and date of approval. Pembrolizumab has the most indications (45) and 40% (18) of these are unique to it. By contrast, tremelimumab, retifanlimab, and toripalimab have fewer indications (2-3) and no non-overlapping indications. Graph of indications by number of ICIs approved. The largest proportion have only one ICI approved (44%) and second-largest 2 drugs approved (29%)
November 12, 2025 at 12:52 AM
hannahrabrams.bsky.social
If a headline ends in a question, the answer is usually no.

Paul & coauthors show that as more drugs have entered the market, price has remained relatively stable.

ascopubs.org/doi/10.1200/...
Table of inflation-adjusted prices for approved ICIs from 2016-2024, ordered by year of approval. First available price has gone up over time and price in Q1 of 2024 has been stable.
November 12, 2025 at 12:52 AM
hannahrabrams.bsky.social
Both drugs have a cleavable linker that allows the MMAE payload (💣) to be cleaved off in the tumor microenvironment.
Diagram of BDC entering tumor microenvironment and cleaving peptide linker, resulting in MMAE affecting tumor Table comparing nectin-4 targeted drug conjugates in development. Highlighted row of BT8009, which has a cleavable linker and MMAE as its conjugated moiety. Source: https://pmc.ncbi.nlm.nih.gov/articles/PMC8224177/
November 10, 2025 at 12:06 AM
hannahrabrams.bsky.social
BDCs build on the concept of ADCs. They have a bicyclic (hence the name) domain that can bind to a given cell surface protein-- in this case, Nectin 4, which is amplified in many solid tumors and the target of enfortumab vedotin (EV).
Figure 3. Structure of Bicycle Toxin Conjugate (BT8009). Source: https://pubs.acs.org/doi/10.1021/acs.jmedchem.2c00065
November 10, 2025 at 12:06 AM
hannahrabrams.bsky.social
🚲⚠️ Look out, CAR T, there's another transportation-named cancer treatment class in town! ⚠️🚲

First-in-human trial of bicycle-drug conjugates (BDCs) hit JCO this week. So what are they, and how are they different than antibody-drug conjugates (ADCs)?

#OncSky #MedSky
Diagram comparing antibody-drug conjugates (ADC) to bicycle-drug conjugates (BDC). On the left, the ADC is larger and labelled as ~150 kDa. On the right, the BDC is small and labelled as ~4 kDA. Both have a 'bomb' emoji representing the chemotherapeutic payload attached to them. The BDC has a lock to represent the cleavable linker.
November 10, 2025 at 12:06 AM
hannahrabrams.bsky.social
Both drugs have a cleavable linker that allows the MMAE payload (💣) to be cleaved off in the tumor microenvironment.
Diagram of BDC entering tumor microenvironment and cleaving peptide linker, resulting in MMAE affecting tumor Table comparing nectin-4 targeted drug conjugates in development. Highlighted row of BT8009, which has a cleavable linker and MMAE as its conjugated moiety. Source: https://pmc.ncbi.nlm.nih.gov/articles/PMC8224177/
November 10, 2025 at 12:05 AM
hannahrabrams.bsky.social
BDCs build on the concept of ADCs. They have a bicyclic (hence the name) domain that can bind to a given cell surface protein-- in this case, Nectin 4, which is amplified in many solid tumors and the target of enfortumab vedotin (EV).
Figure 3. Structure of Bicycle Toxin Conjugate (BT8009). Source: https://pubs.acs.org/doi/10.1021/acs.jmedchem.2c00065
November 10, 2025 at 12:05 AM
hannahrabrams.bsky.social
🚲⚠️ Look out, CAR T, there's another transportation-named cancer treatment class in town! ⚠️🚲

First-in-human trial of bicycle-drug conjugates (BDCs) hit JCO this week. So what are they, and how are they different than antibody-drug conjugates (ADCs)?

#OncSky #MedSky
Diagram comparing antibody-drug conjugates (ADC) to bicycle-drug conjugates (BDC). On the left, the ADC is larger and labelled as ~150 kDa. On the right, the BDC is small and labelled as ~4 kDA. Both have a 'bomb' emoji representing the chemotherapeutic payload attached to them. The BDC has a lock to represent the cleavable linker.
November 10, 2025 at 12:05 AM
hannahrabrams.bsky.social
Interesting article out this week from Dr. Bishal Gyawali, arguing that we should use more superiority designs for treatment de-escalation studies in cancer.

"The burden of proof in medicine lies on more treatment."

academic.oup.com/jnci/advance...
However, SONIA trial7 proves that there is a far more efficient way to answer these questions, by running a superiority rather than a non-inferiority trial. The burden of proof in medicine lies on more treatment. Primum non nocere dictates that the burden of proof should lie on doing something or doing more, rather than on doing nothing or doing less. In other words, if the standard of care is no treatment, use of a new drug should prove to be better than doing nothing; if the standard of care is to use one drug, combination treatment should prove to be better than the single drug.
November 7, 2025 at 10:31 PM
hannahrabrams.bsky.social
- Our voices will never be louder or more abundant than all that information. Our goal is to build relationships that allow people to trust us in doing the best for us.
Slide saying: 4. We can't out compete misinformation. We have to out-relate it.
November 6, 2025 at 10:37 PM
hannahrabrams.bsky.social
- In the world where I can type a diagnosis into an LLM and get targeted marketing across several social media platforms, an oncologist's job isn't just to transmit information. It's helping people curate what info they're processing from the rest of the world about their cancer.
Slide stating: 3. Our task is no longer to transmit knowledge but to co-curate meaning in a noisy information environment.
November 6, 2025 at 10:37 PM
hannahrabrams.bsky.social
- In the modern era, being a doctor doesn't automatically earn trust (and often the opposite, induces some healthy skepticism)
Speaker with slide stating: Trust is now relational capital --not professional capital.
November 6, 2025 at 10:37 PM
hannahrabrams.bsky.social
Among many great talks today at @fredhutch.org HICOR Value in Cancer Care Summit, one that' I'm still thinking about on is this one re: trust in oncology, from Dr. Tony Back. His points (1/5):
Man gesturing to slide that reads, "Trust is now relational capital -- not professional capital." Slide that reads: "3. Our task is no longer to transmit knowledge but to co-curate meaning in a noisy information environment." Slide that reads: "4. We can't out-compete misinformation. We have to out-relate it."
November 6, 2025 at 10:37 PM