๐ฌ Melanoma Intelligence Digest โ Week 2026-W42
Chronic immune-related adverse events after adjuvant anti-PD-1 melanoma therapy persisted in 34% of patients at 5.5 years, poliosis emerged as a marker of durable checkpoint benefit in 741 patients, and real-world data in 580 patients aged 75+ supported adjuvant therapy with 5-year melanoma-specific survival of 74% vs 60%.
โก TL;DR
๐ฆ Twitter / ๐ฆ Bluesky
- Update: chronic immune-related adverse events after adjuvant anti-PD-1 melanoma therapy persist at 5.5 years, pushing clinicians to question adjuvant duration โ 48% of 303 melanoma patients who received adjuvant anti-PD-1 developed chronic immune-related adverse events
Full details below โ
๐ What's Being Published
Poliosis predicts durable response to immune-checkpoint blockade in metastatic melanoma: a multicenter cohort study. โ Brekner Annika et al., J Immunother Cancer
In a retrospective multicenter cohort of 741 patients with non-resectable stage III/IV melanoma treated with ICB between April 2012 and August 2025, 221 developed cutaneous irAE within a median follow-up of 48 months. Tumor recurrence occurred in 17.2% of patients with poliosis, 57.7% with vitiligo and 60.0% with other cutaneous irAE. Poliosis was associated with ORR of 93.1% vs 58.8% vs 38.8% (p<0.001) and PFS of not reached vs 31 months vs 9 months (p<0.001). Poliosis emerged late, at a median of 6 months, and the improved tumor control held regardless of initial tumor burden, treatment duration or melanoma brain metastases.
Clinical bottom line: Clinicians treating metastatic melanoma should record hair whitening on checkpoint blockade. It is a free, late-emerging clinical marker of durable benefit that may support discussions about stopping therapy. The data are retrospective, so it is not yet a basis for protocolized treatment cessation.
Impact of adjuvant therapy on melanoma-specific survival in elderly patients with high-risk resected melanoma: a multicenter international retrospective study. โ Boutros Andrea et al., J Immunother Cancer
Among 580 patients aged โฅ75 years (median age 79 years; range, 75-99) with resected stage III-IV melanoma, 222 (38%) received adjuvant therapy and 358 (62%) underwent observation. Adjuvant therapy was associated with 5-year MSS of 74% vs 60% (adjusted HR 0.58; 95% CI 0.40 to 0.85; p=0.005), OS adjusted HR 0.66 (95% CI 0.51 to 0.87; p=0.002) and RFS HR 0.57 (95% CI 0.45 to 0.72; p<0.0001). Among observed patients, 69% recurred, yet only 47% of those received first-line systemic therapy. Observed patients who stayed recurrence-free or received systemic therapy after recurrence had MSS comparable to adjuvant-treated patients (HR 0.83; 95% CI 0.55 to 1.25).
Clinical bottom line: Fit patients aged 75 and older with resected stage III-IV melanoma should be offered adjuvant anti-PD-1 or BRAF/MEK therapy. Its survival advantage appears to stem largely from ensuring timely treatment, because more than half of observed patients who recurred never received first-line systemic therapy.
๐ Every paper considered this week, including the ones that didnโt make it โ
๐ฃ๏ธ What People Were Discussing This Week
How these are chosen: filtered for clinical relevance, grouped by the new development discussed, then ordered by engagement (bookmarks and reposts).
Update: chronic immune-related adverse events after adjuvant anti-PD-1 melanoma therapy persist at 5.5 years, pushing clinicians to question adjuvant duration
Nearly half of melanoma patients given adjuvant anti-PD-1 developed chronic immune-related adverse events. A substantial share still had ongoing toxicity at long-term follow-up, and hypothyroidism, arthritis and dermatitis frequently persisted. Later onset and corticosteroid-treated events were associated with chronicity. Clinicians are asking whether patients who achieve pCR after neoadjuvant therapy need adjuvant immunotherapy at all, and if so, for how long.
Key numbers: Chronic irAEs 48% of 303 ยท Toxicity ongoing at 5.5 years 34% ยท Hypothyroidism persisted 93% ยท Arthritis persisted 68% ยท Dermatitis persisted 76%
Discussed by: @dr_yakupergun, @JAMANetworkOpen
Sources cited: JAMA Network
๐ฅ Top Ranked Tweets
Gene Therapy Edge @genetherapyedge
Interpath-001: 1,137 patients, resected stage IIB-IV melanoma, vaccine plus pembrolizumab vs pembrolizumab alone. RFS and DMFS endpoints met at interim. Hazard ratios not yet public. ESMO talk Oct 24. #Melanoma #Immunotherapy https://t.co/5QS4QVwa18
๐ 0 โค๏ธ 0 ๐ 0 ๐ 26
OncLive.com @OncLive
๐ค"With the approval of [vusolimogene oderparepvec] plus nivolumab, [the combination] will fit into the post-PD-1 immunotherapyโrefractory space," said Michael Wong, MD, PhD, FRPC, on the @FDA approval in melanoma.
Watch here โก๏ธ: https://t.co/wyph2OHAgN
#melsm #oncology
๐ 3 โค๏ธ 5 ๐ 2 ๐ 1,847
CancerNetworkยฎ @CancerNetwrk
The FDA has approved RP1 plus nivolumab for adults with advanced melanoma who progressed on antiโPD-1 therapy.
https://t.co/dG6guX2Vo9
#melsm #oncology | @FDAOncology https://t.co/zyCvU0aQJe
๐ 1 โค๏ธ 2 ๐ 0 ๐ 642
Yakup Ergรผn @dr_yakupergun
The price of immunotherapy can sometimes continue to be paid for years after treatment ends. Therefore, we need to better define the boundaries of adjuvant IO in particular. Does a patient who achieves pCR really need adjuvant IO? If so, for how long?
In this study, 48% of 303 patients with melanoma who received adjuvant antiโPD-1 developed chronic IRAEs. More strikingly, after a mF/U of 5.5 years, toxicity was still ongoing in 34% of pts. Hypothyroidism persisted in 93%, arthritis in 68%, and dermatitis in 76%.
While trying to increase the chance of cure, we also need to consider the possibility of leaving a cured patient with toxicity that may persist for years.
https://t.co/LhygqMBS66
๐ 16 โค๏ธ 45 ๐ 9 ๐ 3,230
Solange Peters @peters_solange
RT @changmyung1981: ๐งฌ๐ฅ Neoadjuvant immunotherapy is changing melanoma treatmentโand 5-year data now suggest that the tumorโs pathological response after only a short course of therapy may tell us who is already close to cured and who still needs a new strategy.
A major 2026 Nature Medicine analysis from the International Neoadjuvant Melanoma Consortium (INMC) pooled:
1,038 PATIENTS
26 international centers
resectable stage IIIBโIV melanoma
trial + real-world settings
making this one of the largest long-term datasets of neoadjuvant therapy in melanoma. s41591-026-04677-z
The central finding is remarkably clear:
PATHOLOGICAL RESPONSE AFTER NEOADJUVANT ICI
is strongly associated with
LONG-TERM SURVIVAL.
And for patients achieving a:
MAJOR PATHOLOGICAL RESPONSE โ MPR
5-year overall survival approaches:
98%. s41591-026-04677-z
๐ฌ What did the study analyze?
The cohort included:
Neoadjuvant ICI โ 735 patients
BRAF/MEK inhibitors โ 119
ICI + targeted therapy โ 184
Within the ICI group:
PD-1 alone โ 242
PD-1 + another immuno-oncology agent โ 489
including:
PD-1 + CTLA-4 โ 386
PD-1 + LAG3 โ 58.
Median neoadjuvant treatment lasted only around:
7โ9 WEEKS. s41591-026-04677-z
The cohort was overwhelmingly stage III (96.6%), with 49.1% harboring a BRAF V600 mutation. s41591-026-04677-z
๐ฏ First major result: combination ICI produces more MPR
MPR was defined as:
โค10% residual viable tumor
including:
pCR = no viable tumor
near-pCR = >0โ10% viable tumor. s41591-026-04677-z
Among ICI-treated patients:
PD-1 alone โ MPR 49.5%
versus
PD-1 + IO โ MPR 61.4%.
Pathological nonresponse also decreased:
PD-1 alone โ 37.0%
PD-1 + IO โ 27.5%. s41591-026-04677-z
Among specific combinations:
PD-1 + CTLA-4 โ MPR 62.5%
PD-1 + LAG3 โ 63.2%. s41591-026-04677-z
So combination therapy primarily appears to increase the probability of reaching the favorable pathological-response state.
โณ And this difference persists at FIVE YEARS
The KaplanโMeier curves in Figure 1, page 6 show clear separation between treatment groups.
For recurrence-free survival:
5-year RFS
PD-1 alone โ 61.0%
PD-1 + IO โ 73.9%
and for overall survival:
5-year OS
PD-1 alone โ 83.3%
PD-1 + IO โ 87.5%. s41591-026-04677-z
Within combination therapy:
PD-1 + CTLA-4 โ 5-y RFS 72.8%
PD-1 + LAG3 โ 78.7%
although the LAG3 cohort was much smaller and its confidence interval was correspondingly wide. s41591-026-04677-z
๐คฏ But treatment regimen becomes much less important once MPR is achieved
This may be the most important result.
Among patients achieving MPR after neoadjuvant ICI:
5-year OS = 98.0%.
Broken down by treatment:
PD-1 alone โ 98.8%
PD-1 + IO โ 97.9%. s41591-026-04677-z
In other words:
HOW YOU GET TO MPR
may matter less than
WHETHER YOU GET TO MPR.
The authors interpret this as evidence that the survival benefit associated with MPR is highly durable across different ICI regimens. s41591-026-04677-z
๐ The contrast with pathological NONRESPONSE is enormous
Figure 2 on page 7 may be the most clinically important figure in the paper.
After neoadjuvant ICI:
5-year RFS
pCR โ 92.7%
near-pCR โ 86.4%
pPR โ 73.0%
pNR โ 35.2%
s41591-026-04677-z
And:
5-year OS
pCR โ 98.4%
near-pCR โ 96.6%
pPR โ 84.1%
pNR โ 71.6%.
s41591-026-04677-z
This is extraordinary biological stratification after only weeks of therapy.
Conceptually:
NEOADJUVANT ICI
โ
surgery
โ
PATHOLOGY
โ๏ธใใใใใใใใใใใโ๏ธ
MPR pNR
โค10% viable tumorใใใ >50% viable tumor
โใใใใใใใใใใใ โ
5-y RFS โ 91% 35%
โใใใใใใใใใใใ โ
5-y OS โ 98% 72%
s41591-026-04677-z
๐ง Why might neoadjuvant immunotherapy work so well?
The biological rationale is important.
Giving ICI while the tumor and tumor-draining lymph nodes remain intact exposes the immune system to:
THE FULL TUMOR ANTIGENIC LANDSCAPE
during checkpoint blockade.
That may generate broader and stronger antitumor T-cell responses than treating only microscopic residual disease after surgery. s41591-026-04677-z
So instead of:
surgery โ immunotherapy
the strategy becomes:
IMMUNOTHERAPY โ immune priming โ surgery.
This is not merely moving the same treatment earlier.
It potentially changes the immunological context in which the drug operates.
๐ฅ Then comes an extremely important clinical question:
If a patient already has MPR, do they need MORE immunotherapy after surgery?
The dataset suggests:
MAYBE NOT.
Among ICI-treated patients with MPR:
no adjuvant therapy โ 5-y RFS 90.6%
versus
adjuvant ICI โ 90.8%.
Five-year OS was similarly:
no adjuvant โ 98.4%
adjuvant ICI โ 97.6%. s41591-026-04677-z
Essentially:
NO VISIBLE ADDITIONAL BENEFIT.
The same pattern was observed in pPR patients, although the numbers were smaller. s41591-026-04677-z
This supports a powerful response-adapted treatment concept:
RESPONDER โ DE-ESCALATE
NONRESPONDER โ ESCALATE / CHANGE STRATEGY
But because adjuvant therapy was not randomized, this result requires prospective confirmation. The authors explicitly caution about bias by indication. s41591-026-04677-z
๐จ pNR is therefore the major unmet need
Patients with:
>50% residual viable tumor
after neoadjuvant ICI had only:
35.2% 5-year RFS.
These are the patients in whom additional therapy is most urgently needed.
In pNR patients, adjuvant ICI showed a numerical improvement:
no adjuvant โ 5-y RFS 25.6%
adjuvant ICI โ 38.3%
and:
no adjuvant โ 5-y OS 61.9%
adjuvant ICI โ 75.6%.
s41591-026-04677-z
But again:
THIS WAS NOT RANDOMIZED.
So it should be viewed as hypothesis-generating rather than proof of adjuvant benefit.
๐งฌ What about BRAF-mutant melanoma?
This is particularly interesting.
Among ICI-treated patients:
BRAF V600-mutant
โ MPR 46.3%
versus
BRAF WT
โ 55.6%.
And within BRAF V600-mutant disease:
PD-1 alone โ MPR 35.6%
PD-1 + IO โ 51.0%.
s41591-026-04677-z
Five-year RFS after ICI was:
WT/non-V600 โ 75.0%
BRAF V600 โ 59.2%.
But the 5-year OS difference was considerably smaller:
88.8% vs 81.8%. s41591-026-04677-z
So BRAF mutation does not eliminate neoadjuvant ICI activity, but it identifies a population with lower pathological-response and recurrence outcomes in this cohort.
๐ Why not just use BRAF/MEK inhibitors before surgery?
The long-term data are striking.
5-year RFS
ICI โ 70.6%
BRAF/MEKi โ 37.4%
and:
5-year OS
ICI โ 86.5%
BRAF/MEKi โ 69.8%.
s41591-026-04677-z
Even patients achieving MPR with BRAF/MEKi had:
5-y RFS = 57.4%
versus:
91.2% with ICI-associated MPR. s41591-026-04677-z
That is biologically fascinating.
MPR IS NOT THE SAME BIOLOGICAL STATE ACROSS THERAPIES.
Tumor disappearance after targeted therapy may not generate the same durable immune memory as tumor clearance under checkpoint blockade.
The authors therefore conclude that neoadjuvant BRAF/MEKi can largely be abandoned in this setting. s41591-026-04677-z
๐งฌ What about ICI + targeted therapy?
The numbers initially look impressive:
5-year RFS = 76.4%
5-year OS = 83.9%.
s41591-026-04677-z
But this group contained:
7 different regimens
in relatively small numbers.
The authors therefore explicitly caution against drawing firm comparative conclusions. s41591-026-04677-z
This is an important limitation of the pooled dataset.
๐ฆถ ACRAL MELANOMA remains difficult
Among ICI-treated acral melanoma patients:
MPR = 38.1%.
At 5 years:
EFS โ 26.0%
RFS โ 28.5%
OS โ 73.1%.
s41591-026-04677-z
The KaplanโMeier curves in Extended Data Figure 2, page 17 show particularly striking divergence between acral and non-acral melanoma: 5-year RFS falls to 28.5% in acral disease versus 72.5% in non-acral melanoma. s41591-026-04677-z
The cohort is tiny, howeverโonly 26 ICI-treated acral patientsโso dedicated prospective studies are required.
๐งญ What about IN-TRANSIT metastases?
Interestingly, neoadjuvant ICI remained active.
MPR:
nodal disease alone โ 58.2%
in-transit alone โ 66.7%
nodal + in-transit โ 51.1%.
s41591-026-04677-z
And 5-year OS for patients with in-transit metastases alone was:
94.1%. s41591-026-04677-z
Again, numbers are small, but this supports investigation beyond conventional nodal-only disease.
๐ Does this work outside clinical trials?
This is another important feature of the study.
Of the 1,038 patients:
283 โ 27.3%
were treated outside trials. s41591-026-04677-z
Pathological response was remarkably similar:
Trial MPR โ 58.9%
Nontrial โ 55.6%.
s41591-026-04677-z
But long-term disease control was somewhat worse outside trials.
Extended Data Figure 3 on page 18 shows:
5-y EFS
trial โ 71.1%
nontrial โ 61.9%
5-y RFS
trial โ 72.6%
nontrial โ 64.7%
while:
5-y OS
trial โ 86.9%
nontrial โ 85.4%.
s41591-026-04677-z
So efficacy appears transferableโbut implementation quality may matter.
๐ฅ Could SURGERY eventually be reduced?
This is where neoadjuvant therapy becomes even more disruptive.
If a patient achieves MPR in an:
INDEX LYMPH NODE
do we really need complete therapeutic lymph-node dissection?
The authors point to PRADO data showing excellent outcomes among selected index-node MPR patients who underwent:
NO FURTHER SURGERY
and
NO ADJUVANT THERAPY.
s41591-026-04677-z
But the present study does not establish that surgery can routinely be omitted.
Ongoing trials including:
MSLT-3
OMIT
are addressing this directly. s41591-026-04677-z
๐ค And eventually we may want to predict MPR BEFORE treatment
The current paradigm is:
ICI
โ surgery
โ pathology
โ identify responder/nonresponder.
The next paradigm could be:
BASELINE MULTI-OMICS
IMAGING
ctDNA
CLINICAL FEATURES
โ
AI / predictive model
โ
PREDICT MPR BEFORE THERAPY
Then:
predicted responder
โ standard short-course ICI
โ de-escalation
while:
predicted resistant
โ intensified / experimental therapy.
The paper highlights the ongoing NeoIRENIE trial, which is already testing such a predictive allocation strategy. s41591-026-04677-z
๐ฅ This may be the deeper transformation
Traditionally melanoma treatment looks like:
STAGE
โ
surgery
โ
fixed adjuvant therapy.
Neoadjuvant immunotherapy creates something fundamentally different:
TREAT
โ
OBSERVE BIOLOGICAL RESPONSE
โ
ADAPT WHAT COMES NEXT.
Pathology becomes more than a diagnostic endpoint.
It becomes:
AN IN-VIVO FUNCTIONAL ASSAY OF TUMORโIMMUNE BIOLOGY.
And that may be why the result is so powerful.
After only weeks of treatment:
pCR โ 98.4% 5-year OS
versus
pNR โ 71.6%. s41591-026-04677-z
The tumor itself tells us whether the therapy worked biologically.
โ ๏ธ Important limitations
This is a pooled analysis, not a randomized comparison of all treatment strategies.
Treatment groups differed in baseline characteristics and follow-up. The BRAF/MEKi group, for example, had greater tumor burden and more stage IIIC disease. s41591-026-04677-z
The median follow-up for the entire cohort was only:
2.6 YEARS
despite reporting KaplanโMeier 5-year estimates, because some older cohorts provide the long-term tail while many newer patients have shorter follow-up. s41591-026-04677-z
Other limitations include:
small subgroup cohorts
no central pathology review
no central imaging review
heterogeneous ICI+TT
๐ 6 โค๏ธ 19 ๐ 7 ๐ 973
Journal of Clinical Oncology @JCO_ASCO
COMBI-I: Long-Term Overall Survival with Spartalizumab Plus Dabrafenib and Trametinib in BRAF V600โMutant Advanced Melanoma.
Read the full article. https://t.co/NPaXIzzmS8
#melsm
๐ 4 โค๏ธ 5 ๐ 1 ๐ 1,974
Journal of Clinical Oncology @JCO_ASCO
Intismeran Autogene Plus Pembrolizumab Versus Pembrolizumab Alone in High-Risk Resected Melanoma: 5-Year Update of the Randomized Phase IIb KEYNOTE-942 Study. Co-authored by @AdnanKhattakOnc.
Read the full article. https://t.co/AL3Q4QV99o
#melsm
๐ 3 โค๏ธ 3 ๐ 1 ๐ 2,219
OncLive.com @OncLive
An @FDA panel vote that backed RP1/nivolumab in advanced melanoma was not unanimous, with discussions citing no control arm and ORRs that fell once injected-only lesions were excluded.
With the debate and recent FDA approval, should the regimen be adopted into practice? #melsm
๐ 2 โค๏ธ 4 ๐ 1 ๐ 1,292
OncLive.com @OncLive
๐ฅ10 to 3 in favor of RP1/Nivolumab in advanced melanoma was the result of a recent FDA CTGTAC vote supporting the data as evaluable and clinically meaningful for the combination in this setting.
Read more here โก๏ธ: https://t.co/hiBWY9VvmX
#melsm #Oncology
๐ 1 โค๏ธ 3 ๐ 0 ๐ 1,447
Society for Immunotherapy of Cancer @sitcancer
RT @Medicenna1: $MDNAF Topline data from the Phase 1/2 study of MDNA11 in checkpoint resistant cancers has been selected for an oral presentation at #SITC26 โ one of the Top 150 abstracts.
Abstract #517 | Oral Session 3 | Sat, Nov 7, 4:30 p.m. MST
https://t.co/0CXLA2awUI https://t.co/JHoMjV0bxe
๐ 0 โค๏ธ 21 ๐ 5 ๐ 677
Medicenna Therapeutics @Medicenna1
$MDNAF Topline data from the Phase 1/2 study of MDNA11 in checkpoint resistant cancers has been selected for an oral presentation at #SITC26 โ one of the Top 150 abstracts.
Abstract #517 | Oral Session 3 | Sat, Nov 7, 4:30 p.m. MST
https://t.co/0CXLA2awUI https://t.co/JHoMjV0bxe
๐ 0 โค๏ธ 21 ๐ 5 ๐ 677
Society for Immunotherapy of Cancer @sitcancer
RT @Gantosj: $MDNAF $MDNA.TO Top 150๐
Oral presentation! #SITC26
MDNA11 Demonstrates Clinical Activity as Monotherapy and in Combination with Pembrolizumab in Biomarker-Enriched Gastrointestinal Malignancies, Including ICI-Refractory Disease
https://t.co/A0PzyzJjsn
๐ 0 โค๏ธ 21 ๐ 5 ๐ 1,727
Society for Immunotherapy of Cancer @sitcancer
RT @MarcJacksonLA: $MDNAF
Medicenna Announces Oral Presentation of Topline Data from the Phase 1/2 ABILITY-1 Study of MDNA11 at SITC 2026
- The Phase 1/2 ABILITY-1 trial evaluates MDNA11 in expansion cohorts of patients with advanced and metastatic gastrointestinal, endometrial and melanoma cancers.
- Abstract 517 was selected among the meeting's Top 150 Abstracts.
- Its title describes clinical activity in gastrointestinal malignancies selected using biological markers, with MDNA11 alone and with pembrolizumab, including disease resistant to immune checkpoint inhibitors.
- The presentation is scheduled for 4:30โ4:45 p.m. MST in Oral Session 3 at the Phoenix Convention Center, Room 301D, in Phoenix, Arizona.
https://t.co/mWF4xAJBTN
๐ 0 โค๏ธ 10 ๐ 3 ๐ 619
OncLive.com @OncLive
Daraxonrasib wasn't the only notable @FDA #oncology approval in August. Revisit the key regulatory decisions from the month covering melanoma, multiple myeloma, gastric cancer, and polycythemia vera #melsm #mmsm #mpnsm
https://t.co/RURte53v1I
๐ 0 โค๏ธ 6 ๐ 4 ๐ 1,258
VAXiCAN @VAXiCAN_
Yesterday the therapeutic cancer vaccine stopped being a hypothesis.
Congratulations to the Teams at @Merck @moderna_tx and to Professor Georgina Long of @MelanomaAus as well as all #INTerpath001 investigators ๐
#CancerVaccines #ImmunoOncology #eVLP #Biotech #Oncology https://t.co/wtqJ2mgXgU
๐ 0 โค๏ธ 6 ๐ 0 ๐ 712
Prepared by Rafael Santana-Davila, M.D. Questions or feedback? Just reply to this email.
Disclaimer: This digest is AI-generated. Content may contain errors -- always verify with original sources.
Know a colleague who would find this useful? Just forward them this email.
☕ Not funded by anyone — no sponsors, no pharma. Buy me a coffee if it's useful.