A Major Synergistic Win In Lung Cancer!! High Dose Radiation Plus Immunotherapy:
The combination of SABR and short term immunotherapy demonstrates clear improvement for early lung patients. Massive Implications.
Another week - another great radiation outcome for cancer patients!!
Today, we’ll just begin with a pretty impressive curve out of a study from MD Anderson. This is event free survival after initial treatment for “early” lung cancers. And it demonstrates a big win if you combine immunotherapy and SABR radiation.
So this study compares SABR - high dose quick radiation treatments - 1 - 2 weeks and asked the question: Does adding immunotherapy for ONLY 4 cycles help?
And the answer is pretty clear: A Resounding Yes!
So nivolumab (Opdivo) added significantly to early lung cancer outcomes. And importantly not indefinite courses or a year or two or three!! - just 4 cycles.
(one treatment (480mg IV) monthly for 4 months).
This becomes one of the big wins in radiation for the year - likely the most important lung radiation data for 2023. Today we’ll walk through why this is such an impactful study.
It’s clearly a trial that once again demonstrates the extra benefit of immunotherapy to our cancer fighting arsenal, but it also speaks to the power of SABR and its ability to ramp up the immune system. Today, we’ll look focus on the radiation side of the ledger and I’ll try to explain how it might flip standard of care approaches for many patients with early lung cancer.
The Trial:
The question: Can the addition of immune checkpoint inhibition help across a broad population receiving ablative radiation for radiographic early-stage disease?
The combination of the two treatments is labeled I-SABR: immunotherapy plus stereotactic ablative radiation.
The primary endpoint was 4-year event-free survival (local, regional, or distant recurrence; second primary lung cancer; or death). Analyses were both intention to treat (ITT) and per protocol (little changed between the two analysis).
Per the Bristol Meyers Squibb website - here is brief summary for the mechanism of action as a reminder.
What is the rationale for this combination?
SABR is a strong immunomodulator
Emerging evidence indicates that radiotherapy and immunotherapy might have synergistic effects, particularly when the radiation is given as high biologically effective doses (≥100 Gy) in few fractions (1–10). Ablative radiotherapy can transform tumours into a so-called in situ cancer-specific vaccine by boosting the release of tumour-associated antigens, increasing the expression of PD-L1, better activating tumour-directed T lymphocytes to augment local tumour ablation, and better eliminating occult micrometastatic disease.12,13
And this is really part of the why this is big specifically for radiation. The authors include twelve references pointing to a possible synergistic effect between these types of drugs and SABR. So this isn’t just adding nivolumab to any local treatment (think traditional lower dose radiation OR surgery). Here, within SABR ablative doses in a setting of greater lymphopenia protection than traditional approaches, there is clear rationale for this combination to have synergistic, amplified effects. And this is why this is such a strong radiation trial result.
Pathological confirmation was required here - that is often a comparative discussion point between surgery and radiation series - how many might not be cancer? Well, here, that number should be zero. Here even mediastinal staging was strongly recommended. These are well staged patients.
And the radiation details are important so included here:
The prescribed four-fraction (given once per day) SABR dose was 50 Gy to the planning target volume, with a simultaneous integrated boost to the internal gross tumour volume of 60 Gy for peripheral or specific central lesions if dose–volume constraints for radiation-sensitive organs could be met. For all other central lesions, ten fractions (given once per day) were delivered, with a planning target volume dose of 70 Gy and simultaneous integrated boost to the internal gross tumour volume of 80 Gy if dose–volume constraints could be achieved.22 Ultra-central lesions, defined as within 0·5 cm of critical structures (trachea, tracheobronchial tree, oesophagus, heart, major vessels, and brachial plexus) were excluded.
Eligibility is outlined below - primary or isolated lung recurrent non-small cell lung cancer who were non-surgical or unwilling to undergo surgery. But to me, rather broad entry for a SABR trial which increases the trials impact (up to 7 cm of disease allowed):
Early-stage disease was defined as stage IA–IB (tumour size ≤4 cm, N0M0), stage IIA (tumour size ≤5 cm, N0M0), or stage IIB (tumour size >5 cm and ≤7 cm, N0M0)
Arms seem relatively balanced - volume is in fact higher in the immunotherapy arm at 6.4ml vs 4.2ml in the SABR arm (not significantly different, but the benefit does NOT seem to be related to better cancers.
Results: Here is the abstract summary:
Findings From June 30, 2017, to March 22, 2022, 156 participants were randomly assigned, and 141 participants received assigned therapy. At a median 33 months’ follow-up, I-SABR significantly improved 4-year event-free survival from 53% (95% CI 42–67%) with SABR to 77% (66–91%; per-protocol population, hazard ratio [HR] 0·38; 95% CI 0·19–0·75; p=0·0056; ITT population, HR 0·42; 95% CI 0·22–0·80; p=0·0080). There were no grade 3 or higher adverse events associated with SABR. In the I-SABR group, ten participants (15%) had grade 3 immunologial adverse events related to nivolumab; none had grade 3 pneumonitis or grade 4 or higher toxicity.
Looking closer at toxicity, the main difference was in Grade 2/3 fatigue (9 total patients in the I-SABR arm compared to 1). Looking at the toxicity via shortening of nivolumab cycles: 83% completed 3 or 4 of the intended 4 cycles, 17% did stop early - but those perhaps didn’t know how impactful the difference would be so maybe even a higher percentage than post-presentation of this data.
Just to reiterate the improvements with high dose SABR radiation combined with short course immunotherapy treatment:
Local recurrences were less.
Regional recurrences were less.
Distant recurrences were less.
Second lung primaries were less.
Deaths by study end were less.
Read the discussion: I could copy it but read it from the authors. It will serve you well.
To me, it is very likely the lung radiation trial of 2023 moving the standard for more and more patients from invasive surgical approaches towards non-invasive treatments with ablative radiation paired with short term immunotherapies.
If surgery and radiation were close before, this ups the odds that this new merger of treatments might simply win head to head. And, at a minimum, it will put real pressure on cancer specialists to demonstrate that long courses of immunotherapy are in fact, required in this setting. A potential game changing approach on two levels.
Everything is not lost beyond 5 fractions:
(a fractionation aside)
Although the trial title included the word SABR, if the lesion was ultracentral, they kept the dose pretty darn high and extended the fractionation to 10 fractions - ten fractions.
Today, I often think this approach is almost deemed taboo - to give more fractions in order to increase the distance between good safe treatment and high toxicity. I’ve argued for the past few years that we’ve gone a bit too far in too many cases and the easier and safer answer is simply to give a little on the number of fractions.
And don’t kid yourself, there is real pressure to keep the number of fractions as low as possible. It is THE narrative of current times.
Below is text and a graph of SABR for lung cancer demonstrating toxicity in a published series out of the Netherlands: Ten years outcomes after SABR in central and ultracentral primary lung tumors.
Even here, 8 to 12 fractions were used and, to me, it demonstrates the toxicity that for some reason we, at times, accept as we push on number of fractions. Lung cancer is a deadly disease and we need to push for better, but my point is simple: we are pushing quite aggressively in this direction and to me, it is nice to see balance in this study.
As shown, in the large review, Grade 5 toxicity occurred in 12% of patients - directly from treatment toxicity 12% died - as I said, we are pushing quite aggressively. So here, I like the balance. Five is nice but if you have a question, simply add a few more fractions and decrease toxicity risk.
A second quick example on this important current topic:
FAST-Forward breast trial. Yep, a breast trial but I’ve talked about my broad context filing system approach - might be unique but maybe more should consider it. I like to see things work across a wide spectrum of use cases and when that happens, I give those findings a higher priority.
But back to the example. The trial delivered 26 Gy / 5 fractions and it was deemed non-inferior to 40 Gy / 15 fractions but giving just 3.8% more was - in fact - clinically measured and deemed inferior - increasing long term complication risk by 54% - from 9.9% to 15.4%. This type of difference, a 3.8% dose variance, can exist between different clinics based on different prescription implementations even when using the “same” prescription dose and fractionation.
And then online, I see debates about should we completely abandon 15 fraction approaches for, let’s say, 20% of women - really low risk cases. And, honestly, maybe we have to based on the recent New England Journal presentation of the UK Low Import Trial (on my docket of articles to discuss). But from my perspective, I think if you think you can walk the line between non-inferior and inferior that is - at its widest 3.8% according to that data - then do so. But we should not force physicians and patients along that narrow path - at least in my judgement. And 5 days vs. 5 years is nice and easy but really, is 15 days vs 5 years really that different of an argument. To me there is basically no difference. (Note, I’ll move my aggressive hypofractionation clinical approach one step further in breast cancer in favor of short course approaches based on the New England journal article, but that’s different than arguing the 15 is wrong).
Here - in ultracentral lung disease - where we have seen really bad toxicity in trials attempting to deliver super short course, we backed off. We did, in my assessment, what is proven - back off for safety and first verified that 10 fractions can work. And so in applying this data, if you have question regarding safety, use a few more - it is supported!
Yay!! Previously I’ve written about needing to see the pendulum swing back towards the center and when I see this, I see hope.
Summary: Surgery, VALOR, PACIFIC-4
I’ll leave with this: surgery results and two big current trials. This is a reasonably highly cited article looking at Long-Term Results for Clinical Stage IA Lung Cancer- Comparing Lobectomy and Sublobar Resection
1687 patients and 3 year recurrence rates are in a 15%-20% rate. In the current I-SABR study, (which appears to include a bit higher risk cohort), we achieved 75%-80% in likely a little bit sicker / higher co-morbidity group.
It’s an interesting question to ask then. Surgery or radiation? Or perhaps more precisely today: Surgery or I-SABR?
Looking back, it appears quite likely that there is maybe little difference between SABR (without immunotherapy) and surgery. This has been a topic of debate and it is the rationale for several ongoing trials looking to more accurately define the answer to this question. Talking with patients today, I think this new data should give more and more patients the opportunity for a quick, non-invasive 4-10 fraction approach paired with just 4 cycles of monthly nivolumab. It’s an option that many will favor outright.
And of course we have additional ongoing trials. There is the VALOR trial out of the VA attempting to compare SABR to surgical resection. A great trial design - outstanding work to get it off the ground and up and running. But I do think this current study, unfortunately, makes that very “idealistic” trial far less “idealistic”. After all, I now have pretty significant questions regarding the SABR alone arm as representing the “best” standard approach.
Or consider the PACIFIC-4 trial. A second large multicenter trial that again looks at SABR treatment plus either durvalumab (given for 24 months) or a placebo infusion for 24 months. (An EGFR mutation cohort will receive osimertinib for 3 years following treatment).
Wow. From a patient / clinician perspective, I like trial participation but this would be hard to consent to when 4 months off trial has such good data. I know I “should” support all trials, but 4 months in this study gives pretty clear benefit and it avoids 20 months of further treatment which might or might not add further benefit. Pair that scenario with the opportunity to randomize to 2 years of placebo injections where you wouldn’t get any immunotherapy now looks like a tough sell to me. Again an “idealistic” design slides a bit away from the “idealistic” comparison.
If VALOR and PACIFIC-4 had equipoise previously (which they did), this seems to shift the balance even more towards radiation and immunotherapy as a standard alternative non-invasive approach to address early lung cancers. I think it carries immediate impact today in the clinic and it will affect your interpretation of important trials to come in the future. It seems to shake the foundation of two of our larger, more important ongoing trials in lung cancer. It clearly demonstrates that SABR radiation is an important immunomodulator that can be paired with short term immunotherapy to impact lung cancer outcomes. It even raises the question of broader application of this paired synergistic merger in other disease sites. And for those reasons, this is one trial to know - file it prominently in the mental filing cabinet - somewhere way up in the front.
Kudos the team! Kudos to MD Anderson for moving the needle! We come back next week with another, more controversial article from the same Texas town: a look at the largest published series of proton therapy for the treatment of prostate cancer.
As always, opinion of one. Feel free to reach out or comment.
EXTRA REFERENCE (Excellent clinical overview - published February 2023 JCO):
Stereotactic Body Radiotherapy for the Management of Early-Stage Non–Small-Cell Lung Cancer: A Clinical Overview
https://ascopubs.org/doi/full/10.1200/op.22.00475
Nice write-up. I have no doubt more publications will come out looking at the immunological sub-studies they surely did at MDACC using these patients.
Many oncologists were clamoring for adjuvant osi to become SOC after ADAURA showed a big PFS benefit (but hadn’t reported OS). These results are so impressive, I feel myself falling into the camp that we should start using I-SABR now. If not, I hope PACIFIC-4 is viewed as confirmatory, and 4 months of immunotx becomes standard. Let pharma prove that the additional 20 months of immunotherapy is beneficial compared to the 4 months given in this MDACC trial.
Great write up as usual. I enjoy your content and learn a lot.
I will say I think the correct comparison is not going to be I-SABR vs Surgery, but rather I-SABR vs I-Surgery.
To me this study isn't evidence of a synergistic effect as it's possible immunotherapy has an independent effect for early stage lung.
Just as chemo-immunotherapy is making headway in neoadjuvant treatment for surgery in stage III lung I expect there will be efforts by the surgeons to combine Immunotherapy with resection in this cohort as well.