- Joined
- Oct 4, 2017
- Messages
- 5,018
- Reaction score
- 9,787
Let the man have his pub for his resume. I definitely am in favor of novel xrt and those who push forward with it need to be rewarded.
RO may have a role in this COVID crisis, but we need a clinical trial.....
- Thread starter cancer_doc
- Start date
- Joined
- Apr 21, 2011
- Messages
- 3,622
- Reaction score
- 9,267
I went from "Rock on Khan!" to "Please stop Khan!" on this topic fairly quickly.
It's over. Great try. Appreciate the effort. Maybe the next idea will work out.
It's over. Great try. Appreciate the effort. Maybe the next idea will work out.
Ok, who was reviewer 2? How did they miss this one?? *facepalm*
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
More interesting stuff!
www.ncbi.nlm.nih.gov
I read the paper and questions popped up:
"Patients with clinical findings of sepsis, haemodynamic instability, orotracheal intubation (OTI) before radiotherapy or who did not agree to receive treatment with radiotherapy were excluded."
"Regarding the requirement for orotracheal intubation (OTI), there was a tendency toward a decrease in patients treated with radiotherapy, with 33% intubated patients compared with 58% in the control group, although this difference was not statistically significant (P 0.51)."
Could someone please explain this to me?
The investigators excluded intubated patients from trial participation, yet it appears that 10% of patients receiving RT and 58.6% patients without RT were on mechanical ventilation at baseline? Was it non-invasive ventilation? Needless to say, these are not well matched groups if more than half of the patients need mechanical ventilation in one group while only 10% do so in the other group.
It's odd that they mention a 58% intubation rate in the group without RT, which matches exactly the percentage on "mechanical ventilation" in the table at baseline.
The clinical efficacy of low-dose whole-lung irradiation in moderate-to-severe COVID-19 pneumonia: RTMX-20 trial
This is a paired prospective comparative cohort study with 58 patients, in order to analyze the clinical LD-WLI in patients with moderate or severe COVID19 pneumonia. The results of this study show that the Radiotherapy could be an option to improve the ...
www.ncbi.nlm.nih.gov
I read the paper and questions popped up:
"Patients with clinical findings of sepsis, haemodynamic instability, orotracheal intubation (OTI) before radiotherapy or who did not agree to receive treatment with radiotherapy were excluded."
"Regarding the requirement for orotracheal intubation (OTI), there was a tendency toward a decrease in patients treated with radiotherapy, with 33% intubated patients compared with 58% in the control group, although this difference was not statistically significant (P 0.51)."
Could someone please explain this to me?
The investigators excluded intubated patients from trial participation, yet it appears that 10% of patients receiving RT and 58.6% patients without RT were on mechanical ventilation at baseline? Was it non-invasive ventilation? Needless to say, these are not well matched groups if more than half of the patients need mechanical ventilation in one group while only 10% do so in the other group.
It's odd that they mention a 58% intubation rate in the group without RT, which matches exactly the percentage on "mechanical ventilation" in the table at baseline.
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
- Joined
- Nov 2, 2019
- Messages
- 3,497
- Reaction score
- 14,519
1) I agree with your questions about the RTMX-20 data...it's not clear what's going on there.
2) This Indian paper looks stronger...I wonder what confounding gems are to be had in the details.
3) With the new CDC guidance stating healthcare workers (and ONLY healthcare workers) can isolate for 7 days instead of 10, and some states independently saying 5 days, there might be a day in my near future where I just hop up on the TrueBeam table myself in the middle of a busy clinic to give my lungs a wee bit of the Healing Rays so I can go back to falling on my sword for the kind folks who think Bill Gates invented COVID to slow overpopulation while microchipping everyone.
2) This Indian paper looks stronger...I wonder what confounding gems are to be had in the details.
3) With the new CDC guidance stating healthcare workers (and ONLY healthcare workers) can isolate for 7 days instead of 10, and some states independently saying 5 days, there might be a day in my near future where I just hop up on the TrueBeam table myself in the middle of a busy clinic to give my lungs a wee bit of the Healing Rays so I can go back to falling on my sword for the kind folks who think Bill Gates invented COVID to slow overpopulation while microchipping everyone.
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
Indeed. Did you notice that there is no hypothesis for the randomized trial part? They had a hypothesis for the Phase-I exploratory part, but then they simply went on to randomization without a specific statistical plan / primary endpoint power calculation? Weird...
I think they were assessing if they could prevent people from having to go on intubation. These patients probably got worse and required intubation. "The primary objective of the study was to determine survival in patients treated with LD-WLI; the secondary objectives were to determine if there was a difference in orotracheal intubation..." This whole study is wacky...I don't think they excluded intubated patients except in the RT arm (hence the language orotracheal intubation (OTI) before radiotherapy), they probably wanted no intubated patients prior to randomization, but their randomization efforts failed, so they found a cohort to match to, which probably included intubated (or non invasive mechanically ventilated patients), and that doesn't make sense to me, but neither does this paper. "The control group was recruited during the week (from Monday to Thursday) and paired for sex, age and ARDS classification", doesn't look like they matched for intubation status... Or I am reading too much into it.More interesting stuff!
The clinical efficacy of low-dose whole-lung irradiation in moderate-to-severe COVID-19 pneumonia: RTMX-20 trial
This is a paired prospective comparative cohort study with 58 patients, in order to analyze the clinical LD-WLI in patients with moderate or severe COVID19 pneumonia. The results of this study show that the Radiotherapy could be an option to improve the ...
I read the paper and questions popped up:
"Patients with clinical findings of sepsis, haemodynamic instability, orotracheal intubation (OTI) before radiotherapy or who did not agree to receive treatment with radiotherapy were excluded."
View attachment 346826
"Regarding the requirement for orotracheal intubation (OTI), there was a tendency toward a decrease in patients treated with radiotherapy, with 33% intubated patients compared with 58% in the control group, although this difference was not statistically significant (P 0.51)."
Could someone please explain this to me?
The investigators excluded intubated patients from trial participation, yet it appears that 10% of patients receiving RT and 58.6% patients without RT were on mechanical ventilation at baseline? Was it non-invasive ventilation? Needless to say, these are not well matched groups if more than half of the patients need mechanical ventilation in one group while only 10% do so in the other group.
It's odd that they mention a 58% intubation rate in the group without RT, which matches exactly the percentage on "mechanical ventilation" in the table at baseline.
Last edited:
- Joined
- Nov 2, 2019
- Messages
- 3,497
- Reaction score
- 14,519
Eh OK I guess I'll walk back my "the Indian paper looks stronger" a bit.
Beyond the fact that this looks weirdly rushed (the formatting is messed up on the Green Journal's website, and the PDF of the paper is a pre-proof with even more messed up formatting) - you're correct, there's no stated hypothesis or statistical plan for the second phase.
But:
1) It is unclear WHEN this study took place, which I would argue is hugely important given the introduction and adoption of vaccines and various treatments. If you go to India's clinical trials website, it appears to have opened on June 11th, 2020. There's no date of study completion. The "estimated duration of trial" is 6 months, but it looks like that's just what was submitted with the protocol...and we all know how timelines for these things go. The paper was originally submitted in October 2021. It looks like India started vaccinating in earnest in January of 2021 or so. Did anyone in the trial have any level of vaccination? I guess we'll never know!
2) The timeline is also important for drugs available. They address drugs used for ONLY the patients who died, meaning they have the data and know it's important, but only choose to use it here:
The control patients who died received almost all of the COVID drugs, while the LDRT patients who died received almost none of the drugs. That either means the timelines were different or something about the patients were different. The authors acknowledge this as problematic in the last paragraph of the discussion:
"Remdesivir, tocilizumab, Pirfenidone, Vitamin C and zinc were used for patients in both groups in addition to ‘standard pharmacologic treatment’ according to physician’s discretion. Notably, there were several revisions to guidelines about the best use of these pharmacological drugs throughout the duration in which this trial was conducted. This had resulted in varying number of patients receiving these drugs between the two groups and some patients not receiving the drugs, which may have influenced the outcome."
Uh...yeah. Some high-level science going on over here. Here, let me help the authors out a little bit: your hypothesis is that whole-lung LDRT can be used to improve outcomes in COVID patients. If you want to use this secondary "SpO2/FiO2 (SF) ratio" thing, fine (sounds like using PFS to me, but whatever).
Now, here's the tricky part: if your endpoint is improved SF ratio, and your intervention is LDRT, you better introduce no other interventions which could affect SF ratio, or try to control for them if you do. You do neither of these things. You have demonstrated you have access to this data and use it in one table. You try to hand-wave it away in the last paragraph of the Discussion.
This is bad and you should feel bad.
3) Looking at the baseline patient characteristics, you see this gem:
It seems like the D-Dimer values in the LDRT arm were much lower than the control arm. I can't say "significantly" without seeing the primary data. But the median value is 1.5x higher in the control arm...I'm gonna go out on a limb and say that's significant.
Why does this matter? Oh:
"Our results suggest that higher D-dimer levels in COVID-19 patients are significantly associated with disease progression. Higher D-dimer levels, indicative of hypercoagulability, are found to predict disease severity, composite outcomes and mortality events in both unadjusted and adjusted models of odds ratios and time-to-events hazards ratios."
No timeline, unclear use of vaccines, significantly higher baseline D-Dimer levels in control patients, and uncontrolled/unaccounted for use of various pharmacologic interventions.
By "walk back a bit", I mean "shower this paper with Napalm".
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
Precisely! That is my assumption as well. Not well-matched cohorts.
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
This is so great! Thank you!
Concerning timelines: This paper, which came out this summer, was the Phase-I part with 15 additional patients from the Phase-II randomized part.
Whole lung irradiation as a novel treatment for COVID-19: Interim results of an ongoing phase 2 trial in India - PubMed
As per our initial experience, LDRT appears to be a promising modality of treatment with rapid relief of respiratory distress in selected patients with moderate to severe COVID-19 pneumonia. This translates to early clinical recovery and hospital discharge in the selected patient group.
pubmed.ncbi.nlm.nih.gov
Why did the Green Journal allow for that first publication to go through?
"At the time of writing this manuscript, 25 patients (10 in first phase and 15 in second phase) had received LDRT and there were 8 patients in the control group. The preliminary results of these 25 patients who underwent LDRT are presented in this manuscript."
Is it GCP to publish efficacy results of a randomized trial while that trial is still accruing?
The first paper, with the 25 patients, came out in August but states that treatment of these 25 patients took place between "November 2020 and May 2021".
Thus, the other 36 patients of the randomized part (51 total - 15 in the first paper) were recruited from June 2021 to sometime this autumn?
India suffered its big Delta wave in the first half of 2021:
So, the randomized part of the trial mostly happened while numbers were dropping.
In terms of vaccination, it seems that around 10% had a vaccine when the wave hit.
- Joined
- Nov 2, 2019
- Messages
- 3,497
- Reaction score
- 14,519
*sigh*
I had somehow completely forgotten/placed a mental block that many of the COVID papers are just a mad-dash to publish ANYTHING.
I can't imagine the Green Journal allowing anything like this paper to be published if it studied ANY other topic. For many years, I considered the Green Journal stronger than the American Red Journal, but perhaps times are changing (especially with Sue Yom taking over the RJ).
I'm fine with people continuing to publish the LDRT stuff, because I want the trend of "we did all this work and the results were negative/boring so we didn't publish" to stop. But we should all feel an ethical sense of obligation not to oversell what we're publishing or how we're publishing it (dovetailing on my Timmerman rants here). This paper definitely gives us some information to consider, but good Lord the number of confounders blows my mind.
I know SDN has a huge reach in America but just in case my words can find their way to Germany (Baumann) and India:
I either missed this last night or the journal has been updated since my post, but I see in the abstract that "51 RT-PCR positive COVID-19 patients were recruited between November 2020 and July 2021." This is concerning, especially with what @Palex80 wrote above.
This study would be much stronger with -
1) A precise timeline for ALL patients. November 2020 to July 2021 spans the release of vaccines, the release of REGEN-COV, the release of tocilizumab, etc. Comparing what happened to a COVID patient in July 2021 is NOT directly comparable to what happened in November 2020.
2) With the timeline - I believe the predominant vaccine used in India is Covaxin, which is also a two-dose series. I assume there are others in use as well. So, did any of the patients get any level of vaccine, and what was the temporal relationship between vaccine, diagnosis, LDRT (if in the LDRT arm), and discharge/death.
3) With the timeline - a precise accounting of ALL drugs/treatments used and their temporal relationship as well. This is currently all they write:
"For both the groups, corticosteroids were given for a median of 7 days (Range 5-10 days).16/34 (47%) patients in the LDRT group and 10/17 (58.8%) patients in the control group received Remdesivir. Six(17.6%) patients in the LDRT group and five (29.4%) patients in the control group received Tocilizumab."
Just shy of twice as many (percent, not absolute) control patients received tocilizumab, which wasn't "widely" (using that term loosely) available until late Spring 2021.
4) A more thorough accounting of the baseline characteristics of the patients. In addition to the D-dimer issue, there also appears to be an issue with C-reactive protein ("CRP"), with the median value in the LDRT arm being 66 but the median value in the control arm being 90.
CRP also has significant prognostic implications in COVID. How in the world is the fact that BOTH the D-dimer and CRP levels appear to be ~1.5x higher in the controls and this isn't more thoroughly interrogated, or at least discussed, by the authors?
I mean literally, from the paper I just linked there:
"While the associations between CRP and adverse outcomes were consistent among patients with low and high D-dimer levels, patients with high D-dimer and high CRP have the greatest risk of adverse outcomes."
To properly analyze all this (if possible, given the patient numbers) will require the use of more robust statistics/statisticians. Honestly, per the Methods of this paper, the stats they used seem really "basic". What I mean by "basic" is that I can personally replicate what they did in SPSS, and I am absolutely not a statistician. To account for all these variables requires a higher level (in my opinion).
So -
The "positive" Phase I part happened before vaccines and targeted agents, but after the adoption of steroids. As the second part of the trial happened, there was the release and implementation of vaccines, targeted agents, and the Delta wave. There appears to be a significant difference in baseline characteristics of incredibly important prognostic blood markers that are not accounted for.
Also, weirdly, they have tables for the change in blood markers over time (and state there's no significant change from baseline over time for either group, but do not make a statistical comparison of blood markers at baseline).
There is no table describing the temporal change in D-dimer, despite it being recorded at baseline.
Who reviewed this paper?????
- Joined
- Nov 2, 2019
- Messages
- 3,497
- Reaction score
- 14,519
Also, still annoyed about the lack of clarity regarding the timeline, and found this paper looking at the SF ratio and corticosteroids:
The primary endpoint of the Indian LDRT trial was SF ratio over time. Since it says "for both groups, steroids were given for a median of 7 days", we're left to assume all patients got steroids, yes?
But what was the relationship between steroids and delivering LDRT? Looking at the above graph, it's pretty obvious that if SF ratio is your primary outcome, you ABSOLUTELY need to control for the use/timing of steroids in relation to your radiation intervention.
Is this really in the Green Journal? Or is there some sort of DNS spoofing going on and we're looking at something from a high school science fair?
The primary endpoint of the Indian LDRT trial was SF ratio over time. Since it says "for both groups, steroids were given for a median of 7 days", we're left to assume all patients got steroids, yes?
But what was the relationship between steroids and delivering LDRT? Looking at the above graph, it's pretty obvious that if SF ratio is your primary outcome, you ABSOLUTELY need to control for the use/timing of steroids in relation to your radiation intervention.
Is this really in the Green Journal? Or is there some sort of DNS spoofing going on and we're looking at something from a high school science fair?
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
Indeed. This was also an issue in the Khan-report a couple of months ago. Not really clear how corticosteroids were applied and inbalance in the groups. I wonder why this seems like such a big deal to control (and if not controllable) to adequately report in a clinical trial.
I mean, it's like I would be writing a paper on a clinical trial of post-lumpectomy radiotherapy to the breast, saying that patients with RT had better local control and simply stating:
"Hormonal treatment was used in both patient groups, but don't ask me how many patients got it and for how long."
Hahahaha, the Green Journal seems to have a tendency to absorb these not so well written COVID papers... Perhaps there is indeed an issue in the editorial board?
- Joined
- Nov 2, 2019
- Messages
- 3,497
- Reaction score
- 14,519
I had exactly the same thought. Can you imagine the outrage if we submitted a paper about high-risk prostate patients with different radiation dosing regimens, and then writing "all patients got ADT, either Casodex or Lupron, for a median of 9 months (IQR 4 months - 36 months)", and didn't specify any further? We'd get crucified!
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
- Joined
- Apr 3, 2019
- Messages
- 4,269
- Reaction score
- 9,961
GUESS WHO’S BACK, BACK AGAIN
- Mo Kahn
D
deleted941485
Wow…how timely.
- Joined
- Dec 17, 2007
- Messages
- 3,391
- Reaction score
- 4,409
Low-Dose Whole Lung Irradiation for Treatment of COVID-19 Pneumonia: A Systematic Review and Meta-Analysis - PubMed
Despite a mild improvement in intubation-free days, WLI had no significant effect on patients' overall survival. Currently, we cannot recommend routine use of WLI for the treatment of patients with moderate-to-severe COVID-19.
pubmed.ncbi.nlm.nih.gov
We even have a meta-analysis now!
- Joined
- Oct 24, 2010
- Messages
- 3,512
- Reaction score
- 6,365
- Joined
- Oct 10, 2011
- Messages
- 8,624
- Reaction score
- 10,698
Supporters of RT for COVID - more or less cult-like than proton following?
- Joined
- Apr 3, 2019
- Messages
- 4,269
- Reaction score
- 9,961
WAYYYYYYYYY less culty for COVID RT'ers, speaking purely in terms of venality. Jim Jones needed followers *and* money.
- Joined
- Aug 4, 2020
- Messages
- 1,967
- Reaction score
- 3,354
