Sai kumar Samala MD, MBBS (@SaiKumarSamala_) header
Sai kumar Samala MD, MBBS (@SaiKumarSamala_) profile picture

Radiation Oncologist. Aspiring Physician Scientist Advanced Precision Radiotherapy Research Fellow , UCLA MD Radiation Oncology (AIIMS New Delhi) ,MBBS

Los Angeles
Joined March 2014
Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)Photo by Sai kumar Samala MD, MBBS (@SaiKumarSamala_)
Sai kumar Samala MD, MBBS retweeted
They're calling it the greatest opening paragraph of any recent oncology article
33
274
16
2,288
95,872
Sai kumar Samala MD, MBBS retweeted
Come with me into Varian’s North American training center to learn about the evolution of radiotherapy, which dates back to 1847. #radonc 1/
6
10
1
53
4,114
New #VirtualTumorBoard presents a patient with recurrent #ProstateCancer after radical prostatectomy, covering biochemical recurrence to metastatic disease. 🔗: acsjournals.onlinelibrary.wi… @OncoAlert @AmarUKishan @omgitsyuliaz @KekoaMDPhD @MSteinbergMD @CalaisJeremie #PCAM
1
25
1
40
4,794
Sai kumar Samala MD, MBBS retweeted
Sean Pitroda, whose work has raised concerns about the Badscopal effect, shares a few words of wisdom at #ASTRO26. Including the importance of total metastatic ablation of all visible sites in case irradiation of one tumor signals another unirradiated to grow.
3
17
33
2,667
Sai kumar Samala MD, MBBS retweeted
This is tremendous work from @DrRishabhOnco, @DrNikhilvasudev, @hema_oncologist, @dradhiparora, @aparna96, @RanjitKSMD, @brusabhanu1, @drsameerbakhshi - led & mentored by @batraatulonc. The study harnesses the power of @aiims_newdelhi, an institution that has generated multiple leaders in our field. Editorial forthcoming in @ESMO_Open! Stay tuned!
Proud of @DrRishabhOnco for three years of work on this study—answers a crucial question for LMICs where PARPi access remains a barrier. @neerajaiims @montypal @aparna96 @aiims_newdelhi @DrNikhilvasudev @dradhiparora
2
11
1
40
4,980
Sai kumar Samala MD, MBBS retweeted
🚨🚨 Randomized Trial #ASTRO26 Resected Brain Met fractionated SRS vs single fraction SRS post-op cavity 🚨 • fSRS superior surgical bed control • No diff AEs • fSRS standard of care for resected larger brain met
8
66
4
161
11,184
Sai kumar Samala MD, MBBS retweeted
Most RT-drug combos bet on synergy. Yet synergy is rare (most combos only additive). We're chasing a bar most of oncology can't clear. Our phase 1s don't even measure it. ProjectOptimus for #RadOnc & our editorial in @IJROBP w/Gustav, Ezra & @imrtlee - doi.org/10.1016/j.ijrobp.202…
Must listen and must read!! See also accompanying commentary by Nadeem Riaz @xrtGenomics which should be out soon in @IJROBP #RedJournal #radonc #radbio @finn_corinne @sueyom
2
10
34
3,877
We cover its mechanisms, workflow, and clinical outcomes: durable reductions in VT burden and ICD shocks, but high recurrence and limited long-term data mean it's still investigational, pending randomized trials. Thanks to all coauthors & corresponding authors!
14
Sai kumar Samala MD, MBBS retweeted
The bolus versus weekly cisplatin debate rages on, but fortunately more data dropped this past week with the publication of the ConCERT trial (Concurrent Chemotherapy and External Radiation Therapy) comparing bolus with weekly schedules. Some thoughts on this trial and the cisplatin schedules below. academic.oup.com/jnci/articl…
2
25
1
48
3,681
Sai kumar Samala MD, MBBS retweeted
This paper is incredibly important. The authors asked a simple question: how reproducible is CTCAE v5.0 acute toxicity scoring during head and neck radiotherapy? It turns out that it’s not very reproducible. During a prospective clinical trial (specifics of the trial not included), two experienced H&N radiation oncologists evaluated CTCAE v5 defined dermatitis, oral mucositis, dysphagia and xerostomia once per week, within 24 hours of each other.
Documenting physician reported toxicities can vary between clinicians. This also raises the importance of documenting patient reported outcomes . A very pragmatic study done by TMH Head Neck Oncologist . Congratulations ! @RadOncTMC @shwetabhsinha23 @AnujkumarTvm
2
17
2
44
7,835
Sai kumar Samala MD, MBBS retweeted
After reading a lot of grant proposals, I have a theory for why the valley of death in translational research exists: 1. You can make it exciting to propose doing X-omics or testing role of process Y on a disease, even if its the 10th omics study and the 20th process tested…
3
20
1
210
55,870
Sai kumar Samala MD, MBBS retweeted
A thoughtful virtual tumor board tackles a problem we will see more often: UIR prostate cancer with Decipher 0.94/high risk but Artera low prognostic risk and a negative ST-ADT predictor. It frames the problem very well—but does not resolve it (and unlikely the authors' intention). “Discordant” may be the wrong frame. Decipher is a 22-gene prognostic assay; Artera is a pathology-clinical AI platform with prognostic and predictive components. They estimate different things. A high-risk tumor can still be relatively ADT-insensitive. The key inferential bridge is that high Decipher risk supports ADT because higher baseline risk yields greater absolute benefit if ADT’s relative effect is constant. The authors explicitly acknowledge that assumption. Artera sought to challenge that assumption. So this is not simply Decipher versus Artera. It is (A) a risk model plus an assumed treatment effect versus (B) a treatment-effect model. The outputs should be translated into expected absolute benefit—with uncertainty—not compared as “high” versus “negative.” The Artera result should not be overread. A negative predictive result means no clear average benefit was detected in that subgroup—not proof this patient receives zero benefit. Calling the subgroup “insensitive” is stronger than a probabilistic model supports. The paper advises reviewing the raw Decipher score, not just its category. Yet the Artera treatment prediction is presented as binary. Without its continuous score, distance from threshold, interaction estimate, and CI, adjudication is inherently asymmetric. The reports on pp 2–3 are prognostically worlds apart: 10-year DM 17.4% with Decipher versus 1.3% with Artera. Also missing and key issue: were both assays performed on the same lesion, core, and tissue block? With multifocal disease and 7/12 systematic plus 2/3 targeted cores positive, spatial sampling could create apparent discordance. The case is clinically tilted toward ADT: cT2b, GG3, ~60% pattern 4, ≥50% positive cores, plus substantial recurrence anxiety. A safety-first recommendation is understandable—but this is not a neutral stress test of biomarker-guided de-escalation. To their credit, the authors default to randomized evidence. RTOG 0815 showed 10-year absolute differences of ~8% in biochemical failure, 4% in DM, and 2% in PCSM, without an OS benefit. That is the counseling frame: absolute benefit versus ADT burden. Figure 3 is intuitive but causally hazardous. Population cumulative-incidence curves cannot identify which individuals were “overtreated,” “benefited,” or remained “undertreated.” Those are counterfactual response categories, not observed patient strata. Simon level 1B and guideline recognition support validation, not prospective clinical utility. We still need paired testing, same-block analyses, UIR calibration, decision curves, and ideally randomized evidence that biomarker-guided care improves net benefit (RCTs in progress). Bottom line: excellent paper for separating prognosis from prediction. But the answer is not to pick a winner. Until risk, treatment effect, uncertainty, and preferences are integrated quantitatively, more testing may yield more data—not better decisions.
2
3
10
850
Sai kumar Samala MD, MBBS retweeted
🚀 Our randomized Phase II SRAM trial is now published! 📊 Whole-pelvis #SBRT vs conventional pelvic IMRT for high-risk #ProstateCancer: ✅ Acute Grade ≥2 GI toxicity ⬇️ 39.0% → 8.3% (P<0.0001) ✅ Comparable GU toxicity ✅ Better early QoL ✅ Only 5 weekly fractions vs 38 fractions. Our study extends the evidence to pelvis-inclusive SBRT. Together with ongoing Phase III trials (#PACENODES, #PRIME, #HighFive, #ASCENDESBRT), the momentum toward ultra-hypofractionated treatment for high-risk disease continues to grow. 🚀 @alison_tree @VedangMurthy @nickva1 @indranilsays @TylerSbrt @piet_ost @PBlanchardMD @PaulSargos @APCCC_Lugano @BarbaraJereczek @ESTRO_RT @ASTRO_org @OncoAlert @oncodaily @QuadShotNews @HKSUO_hk @myESMO @DrSpratticus @ChadTangMD @SbrtSean @drmikechao @IJROBP @docpriyamvada redjournal.org/article/S0360…
5
31
3
70
4,775
Sai kumar Samala MD, MBBS retweeted
1/ Local control after lung SBRT is >90%, right? Depends who's counting, and how. A new systematic review (Nourmohammadi et al, Radiother Oncol '26) reviewed 113 studies and found we can't even agree on what "local control" means. 🧵
8
45
5
81
12,369
Sai kumar Samala MD, MBBS retweeted
How to predict PD-L1 expression level without biopsy for your patients? Houston, We Have a Solution. We just published “Deep Learning of CT Imaging Predicts PD-L1 Expression and Immunotherapy Response in Metastatic NSCLC: A Multi-Center Study” in Cancer Letters by team leader @JiaWu_PhD My personalized Share Link: authors.elsevier.com/c/1nIqS… SCENT, an ensemble transformer model, predicts PD-L1 (≥50% vs <50%) from routine CT as a noninvasive virtual biopsy in metastatic NSCLC. • SCENT achieved strong discrimination for PD-L1 status in the MDACC cohort (AUC 0.84) and generalized it to external cohorts (AUC 0.80 Mayo; 0.78 LONESTAR trail). • SCENT-derived PD-L1 stratified immunotherapy outcomes, associated with both PFS (HR 1.49) and OS (HR 1.40), comparable to tissue PD-L1 IHC.
8
40
5
136
14,063