ESMO 2026 and Innovations in Cancer – 18 Key Studies: Tarlatamab, VOLGA, ADCs, Chemotherapy-Free Immunotherapy, and a Warning on MOIO

18 Trials from Madrid: What ESMO 2026 Changed in Cancer Treatment, and What It Did Not

The 2026 congress of the European Society for Medical Oncology (ESMO) took place in Madrid on 23–27 October. The late-breaking abstracts presented in four Presidential Symposia, together with the congress's other major presentations, reopened the debate on standards of care in many areas, from small cell lung cancer to bladder, stomach, biliary tract, endometrial and bone cancer. In this article, we go through the 18 standout trials one by one, with their numbers and their limitations, and assess which ones may change everyday practice and which ones are still too early to call. You can find the congress program on ESMO's official page. Before the congress, we covered the 12 trials set to shape the agenda and the new stomach and esophageal cancer guideline; during the congress, we wrote separate articles on EGFR-mutated lung cancer, KRAS G12C targeted therapies, OptiTROP-Lung06 in PD-L1-negative lung cancer, SANO in esophageal cancer, KEYNOTE-B49 in breast cancer, DEWI in stomach cancer and STARBOARD in melanoma.

Why does this matter?

Together with the annual meeting of the American Society of Clinical Oncology (ASCO), the ESMO congress is one of the two largest cancer congresses in the world. Phase 3 trials presented there are often reflected in drug approvals and guidelines within a few months. The 2026 congress stood out for positive phase 3 results in diseases that have had no new treatment options for a long time (small cell lung cancer, biliary tract cancer and relapsed osteosarcoma).

On the other hand, congress data are first analyses: follow-up is short, some results have not yet been published in a peer-reviewed journal, and some rely on company announcements. That is why, for every result, alongside "What was found?" we also need to ask "How certain is it, who does it apply to, and how big is the absolute difference?" (reading research correctly). In this article, we give the relative effect (risk reduction) and the absolute effect (the difference in months or percentage points) together.

What is changing in cancer treatment after ESMO 2026?

Four main trends stood out at ESMO 2026: antibody–drug conjugates beat chemotherapy in five different cancers; immunotherapy moved into the period before surgery and into chemotherapy-free first-line treatment; next-generation targeted drugs outperformed the older generation; and less treatment proved safe only when backed by evidence. The largest survival gain was achieved with tarlatamab in small cell lung cancer.

The 18 trials in one table

An HR (hazard ratio) below 1 favors the new treatment; above 1, it works against it. MATTERHORN's final overall survival result was first presented at ESMO 2025; it stayed on the agenda in 2026 with its full publication and its entry into guidelines.

Which treatments at ESMO 2026 gained the most survival?

In trials that extended overall survival with statistical significance, the median gain ranged from 2.8 to 8.4 months. The largest gain was seen in extensive-stage small cell lung cancer, where median survival is usually just over 1 year. It is worth remembering that the same gain does not mean the same thing in every disease: in later-line stomach cancer, where median survival is 5 months, a 5-month gain means survival doubles.

I · Early stage: trials aiming to raise the chance of cure

1. INTerpath-001: personalized mRNA therapy in melanoma in phase 3

In 1,137 patients with stage IIB–IV melanoma that had been completely removed by surgery, the mRNA therapy intismeran autogene (mRNA-4157/V940), produced according to each patient's own tumor, was added to pembrolizumab. This treatment is not a preventive vaccine in the classic sense: the removed tumor is sequenced, new tumor-specific antigens (neoantigens) are identified, and a personalized mRNA molecule is produced that directs the immune system toward these targets (cancer vaccines and immunotherapies).

The risk of recurrence or loss of life fell by 42%; in absolute terms, at 2 years about 12 more patients out of every 100 were free of recurrence. Grade 3 or higher treatment-related side effects were similar (14.2% vs 13.4%). However, overall survival data are not yet mature, and the place of this treatment is being debated at a time when immunotherapy before surgery is becoming widespread. The results were made public before the presentation after the congress embargo was broken.

2. VOLGA: treatment before surgery for bladder cancer patients who cannot receive cisplatin

In muscle-invasive bladder cancer, the standard is removal of the bladder after cisplatin-based chemotherapy. However, about half of patients cannot receive cisplatin because of kidney function, age or other illnesses, and they often go straight to surgery. In VOLGA, durvalumab + enfortumab vedotin before surgery and durvalumab after surgery were compared with going straight to surgery in these patients (695 patients in total; tremelimumab was also added in one arm). We have previously written that a similar approach with pembrolizumab received FDA approval.

The two-drug treatment reduced the risk of an event (recurrence, progression or loss of life) by 49% and the risk of loss of life by 42%. In absolute terms, at 2 years the proportion of patients living event-free rose by 19 points and the proportion alive by 9 points. The price is more side effects: grade 3–4 side effects 64% vs 44%; skin rash (49%) and nerve damage (18%) caused by enfortumab vedotin stood out. The three-drug arm (adding tremelimumab) gave no extra benefit and caused more side effects.

3. DASL-HiCaP: darolutamide in very high-risk prostate cancer

In very high-risk prostate cancer that has not spread (risk classification), the standard treatment is long-term androgen deprivation therapy and radiotherapy. In the ANZUP-led DASL-HiCaP trial, 1,106 patients also received 96 weeks of darolutamide or placebo. 85% of patients received radiotherapy as their first treatment and 15% because of recurrence after surgery.

The risk of metastasis or loss of life fell by 58%; in absolute terms, at 4 years the proportion of patients living without metastasis rose by 7 points. The number of patients who developed metastases was 9 vs 46. Side effects were very similar to placebo (hot flashes 66% vs 64%, fatigue 64% vs 57%). Overall survival data are not yet mature; this is an interim analysis result.

4. NATALEE: overall survival at 6 years in early-stage breast cancer

In stage II–III, hormone receptor-positive, HER2-negative early-stage breast cancer, adding 3 years of ribociclib after surgery showed, for the first time at 6 years, a difference in overall survival as well: 91.9% vs 89.7% (absolute difference 2.1 points; HR 0.81). The difference in invasive disease-free survival reached 5.3 points (83.2% vs 77.9%). Treating about 47 patients with 3 years of ribociclib prevents one loss of life over 6 years. The p value is nominal; the benefit is small but widens over time (criticisms of CDK4/6 inhibitors and the current picture).

5. MATTERHORN: durvalumab is here to stay in operable stomach cancer

In operable cancer of the stomach and the gastroesophageal junction (948 patients), adding durvalumab to FLOT chemotherapy before and after surgery had raised pathological complete response from 7.2% to 19.2% and 2-year event-free survival from 58.5% to 67.4%. In the final analysis, the proportion of patients alive at 3 years rose from 61.9% to 68.6% (absolute difference 6.7 points; HR 0.78). This result was first presented at ESMO 2025; in 2026, with its full publication and its entry into the ESMO guideline, the new standard became established. The benefit appears more marked in patients with a good pathological response after surgery (stomach cancer guide).

II · Advanced stage: new drugs and new combinations

6. DeLLphi-305: the result of the year in small cell lung cancer

In extensive-stage small cell lung cancer, the first-line standard is platinum-etoposide + immunotherapy followed by immunotherapy maintenance; however, median survival is about one year. Tarlatamab is a bispecific T-cell engager that links the DLL3 protein on the tumor cell to immune cells. In DeLLphi-305, tarlatamab was added to durvalumab as maintenance therapy in 563 patients whose disease had not progressed after 4 cycles of induction.

The risk of loss of life fell by 58%, and median overall survival was extended by 8.4 months. The response rate was 26.4% vs 4.1%. Cytokine release syndrome occurred in about half of patients, but the great majority of cases were mild; grade 3 occurred in only one patient. Stopping treatment because of tarlatamab remained at 3.2%. The results were published simultaneously in NEJM. We covered the topic in detail in our article on new treatments and an access map for small cell lung cancer.

7. Krascendo 1: new generation vs old generation in KRAS G12C

In previously treated KRAS G12C-mutated lung cancer (338 patients), the second-generation drug divarasib, compared with the approved drugs sotorasib or adagrasib, raised median progression-free survival from 5.7 to 9.8 months (HR 0.58) and overall survival from 12.0 to 17.7 months (HR 0.56). The response rate was 58% vs 32%. The results were published in NEJM. Notably, although divarasib was taken for longer, grade 3 or higher side effects were less common (24.6% vs 34.5%).

8. SAFFRON: a chemotherapy-free option for MET increase after osimertinib

In EGFR-mutated lung cancer, one of the main causes of resistance after osimertinib is a MET increase. In SAFFRON, in 338 patients with high-level MET overexpression or amplification, the tablet combination osimertinib + savolitinib extended progression-free survival (7.6 vs 5.4 months; HR 0.55) and overall survival (17.8 vs 12.9 months; HR 0.68) compared with platinum-pemetrexed chemotherapy. The response rate was 51% vs 27%. Grade 3 or higher side effects were similar (64% vs 63%); however, loss of life due to side effects occurred in 12 patients (7%) in the combination arm and in none in the chemotherapy arm. Selecting the right patient requires a repeat biopsy at the time of progression.

9. CLARITY-Gastric01: a claudin 18.2-targeted conjugate in later lines

Claudin 18.2 is a protein often found on the surface of stomach cancer cells; the first drug targeting it was zolbetuximab. Sonesitatug vedotin (sone-ve), on the other hand, is an antibody–drug conjugate that binds to this protein and carries chemotherapy into the cell. The trial enrolled 594 previously treated patients with claudin 18.2 present in at least 25% of tumor cells; about two-thirds were in the third or a later line.

In the third line and beyond, median overall survival doubled (from 5.0 to 10.0 months); the proportion of patients alive at 12 months rose by 20 points. In all patients, however, the difference in progression-free survival was not significant on independent review (4.2 vs 3.8 months). Nausea, loss of appetite and vomiting were clearly more common; grade 3 or higher side effects, however, were less common (52.9% vs 61.1%).

10. G-HOPE-001: a second conjugate for the same target

In 464 patients in China and Japan who had received at least two lines of treatment and whose tumor was strongly claudin 18.2-positive in 75% or more of cells, arcotatug tavatecan, compared with the physician's choice of chemotherapy, raised median progression-free survival from 3.1 to 7.1 months (HR 0.45) and overall survival from 6.9 to 10.2 months (HR 0.57). The response rate was 37.4% vs 7.7%. The results rely on a company announcement; whether they apply to patients in Europe and Turkey needs to be confirmed.

11. HARMONi-GI1: a PD-1/VEGF bispecific antibody in biliary tract cancer

In advanced biliary tract cancer (cholangiocarcinoma and targeted therapies), the first-line standard is gemcitabine-cisplatin + durvalumab or pembrolizumab. Ivonescimab is a bispecific antibody that targets both PD-1 and the blood vessel growth factor VEGF in a single molecule. In the trial, conducted in China, 682 patients received ivonescimab or durvalumab together with the same chemotherapy.

Against an active and effective standard, the risk of loss of life fell by 25%; the median gain was 2.8 months, and at 18 months the proportion of patients alive was 11.5 points higher. Grade 3–4 side effects were more common. In the HARMONi lung cancer trial, which also included Western patients, the overall survival difference with ivonescimab + chemotherapy had not reached the significance threshold; for this reason, the finding in biliary tract cancer needs to be confirmed in patients outside China.

12. PANKU-Breast01: an EGFR×HER3 conjugate in hormone-positive breast cancer

In HR-positive, HER2-negative metastatic breast cancer that had progressed after hormone therapy and at least one line of chemotherapy (383 patients, China), izalontamab brengitecan (iza-bren) reduced the risk of progression by 59% compared with chemotherapy: median progression-free survival rose from 4.0 to 8.2 months, and the response rate from 20.2% to 40.5%. However, grade 3 or higher treatment-related side effects were clearly more frequent (85.3% vs 65.8%), and overall survival data are not yet mature. There was no comparison with the other conjugates already used in this patient group (new options in HR-positive breast cancer, antibody–drug conjugates in breast cancer).

13. TroFuse-005: the first survival benefit for a conjugate in endometrial cancer

In endometrial cancer (cancer of the lining of the womb), options after platinum chemotherapy and immunotherapy are limited; response rates with doxorubicin or paclitaxel are low. In TroFuse-005, the TROP2-targeted drug sacituzumab tirumotecan (sac-TMT) was compared with the physician's choice of chemotherapy in 776 patients from 31 countries. In about 84% of patients, the tumor's DNA repair system was intact (pMMR).

The risk of loss of life fell by 32% and the risk of progression by 41%; median overall survival was extended by 6.9 months. This is the first phase 3 trial in which a TROP2-targeted conjugate extended overall survival in endometrial cancer. Anemia, mouth sores and eye surface problems were more common; preventive mouth and eye care is important for these side effects.

14. ARTEMIS-011: a B7-H3-targeted conjugate in relapsed osteosarcoma

Osteosarcoma is a bone cancer seen mostly in adolescents and young adults. When it relapses, there are very few effective options and almost no randomized trials. In ARTEMIS-011, conducted in China, 122 patients who had received at least two lines of treatment (median age 21–22; about a quarter were adolescents) received the B7-H3-targeted drug risvutatug rezetecan (ris-rez) or gemcitabine + docetaxel.

Median progression-free survival rose from 1.4 to 5.5 months (HR 0.33); the proportion of patients free of progression at 4 months was 65% vs 33%. The overall survival trend is favorable but not yet significant (HR 0.51), and 17 patients in the control arm later switched to ris-rez. Grade 3 or higher treatment-related side effects (especially low white blood cell counts) were very common (93% vs 83%). It is a small trial with short follow-up; a global trial including patients outside China (EMBOLD Sarcoma-202) is ongoing.

15. LITESPARK-011: survival difference not significant in kidney cancer

In clear cell kidney cancer that had progressed after immunotherapy, belzutifan + lenvatinib had extended progression-free survival compared with cabozantinib and, on the basis of this result, had received FDA approval. In the final analysis presented at ESMO 2026 (747 patients, median follow-up 38.8 months), median overall survival was 33.7 vs 28.6 months; however, the difference did not reach the prespecified statistical threshold (HR 0.85; 95% CI 0.70–1.03). The differences in progression-free survival (14.8 vs 10.8 months) and response rate (53.1% vs 40.2%) were maintained; grade 3 or higher side effects were similar. Conclusion: the combination remains an option, but there is no strong survival evidence for it to replace cabozantinib.

III · Less treatment: when is it safe, and when is it not?

16. KEYNOTE-C93: chemotherapy-free first-line treatment in dMMR endometrial cancer

In about a quarter of endometrial cancers, the DNA repair system is deficient (dMMR); these tumors respond very well to immunotherapy. Today, the first-line standard is chemotherapy + immunotherapy. In KEYNOTE-C93, pembrolizumab alone (once every 6 weeks) was compared with carboplatin + paclitaxel in 299 patients.

Median progression-free survival rose from 8.4 to 38.2 months; the risk of progression fell by 43%. Response rates were similar (64.1% vs 62.1%), but responses with pembrolizumab were far more durable. Grade 3 or higher side effects fell to a third. The overall survival difference was not significant; however, this is expected, because most patients in the chemotherapy arm whose disease progressed later received immunotherapy. An important caveat: in the first 6 months, the disease progressed earlier with pembrolizumab in some patients; new markers are needed to predict who will not respond to immunotherapy alone. The trial compared pembrolizumab only with chemotherapy, not with the chemotherapy + immunotherapy combination.

17. MOIO: is it safe to space out immunotherapy?

For patients responding to immunotherapy, how long and how often the treatment should be given is an important question: less frequent treatment could reduce hospital visits, side effects and costs. In the French MOIO trial, 340 patients who were responding after 6 months of immunotherapy (most with lung, bladder or kidney cancer) were assigned to treatment at standard intervals or once every 12 weeks.

A "not worse" (non-inferiority) result for progression-free survival could not be shown; moreover, the rate of loss of life was twice as high in the spaced-out arm. Grade 3–4 side effects fell only slightly (29.8% vs 34.5%). Enrollment in the trial was stopped following a warning from the independent data monitoring committee. The conclusion is clear: for a responding patient, lengthening the interval between immunotherapy doses on one's own is not recommended. The duration of treatment and any decision to pause should be made together with the oncologist in light of the evidence (immunotherapy side effects and their management).

18. SANO: active surveillance in esophageal cancer

In 309 patients with a clinical complete response to chemoradiotherapy before surgery, the difference in 5-year overall survival between patients whose surgery was postponed and who were closely monitored (198) and those who had standard surgery (111) was not significant (50% vs 54%; HR 1.17). However, with active surveillance, distant metastasis (46% vs 29%) and loss of life due to esophageal cancer (40% vs 30%) were more common. In about a third of patients, the esophagus was preserved permanently. Active surveillance is an option in selected patients who prioritize quality of life and understand the risks, at experienced centers; it is not a standard for everyone.

Five big lessons from ESMO 2026

1. Antibody–drug conjugates have spread to every area of oncology. Five conjugates aimed at four different targets (TROP2, EGFR×HER3, claudin 18.2 and B7-H3) beat chemotherapy in phase 3 trials in endometrial, breast, stomach and bone cancer. However, these drugs are "targeted chemotherapy": anemia, low white blood cell counts, mouth sores, and eye and lung side effects need to be monitored closely.
2. Immunotherapy is moving earlier and into chemotherapy-free use. VOLGA and MATTERHORN showed the value of immunotherapy before and after surgery, INTerpath-001 the value of personalized mRNA therapy, and KEYNOTE-C93 the value of chemotherapy-free treatment in correctly selected patients.
3. The right patient comes before the right drug. Without testing for MET increase in SAFFRON, claudin 18.2 in CLARITY-Gastric01 and G-HOPE-001, dMMR in KEYNOTE-C93 and KRAS G12C in Krascendo 1, it is not possible to benefit from these results. Comprehensive biomarker testing at diagnosis and at the time of progression is now a necessity.
4. Less treatment is safe only when backed by evidence. While KEYNOTE-C93 and SANO showed that less treatment is possible in selected patients, MOIO gave a warning that spacing out immunotherapy may shorten survival.
5. Relative effect, absolute effect and overall survival must be read together. In LITESPARK-011 and KEYNOTE-C93, the overall survival difference was not significant; in NATALEE, the absolute difference is 2.1 points. The fact that a substantial share of the trials were conducted only in China also calls for caution in generalizing the results to other populations.

Which trials will change everyday practice?

Practical tips for patients and their families

• Do not change your treatment on your own: Although the congress results are encouraging, most are not yet approved. Especially if you are receiving immunotherapy, always make any decision to lengthen the intervals or take a break together with your oncologist.
• Ask about your biomarker tests: EGFR, KRAS and MET in lung cancer; claudin 18.2, HER2 and MSI/MMR in stomach cancer; and MMR status in endometrial cancer are prerequisites for access to new treatments.
• Look into clinical trials: For these drugs, which are not licensed in Turkey, clinical trials are often the safest route to access.
• Ask about the absolute benefit and the side effects: "In my situation, how much extra time on average, or what percentage of extra chance, does this treatment offer, and at what cost?" is the key question for making a decision.
• Do not hesitate to get a second opinion when making a decision.

Frequently asked questions

What were the most important results of ESMO 2026?

The strongest results came with tarlatamab + durvalumab in small cell lung cancer (overall survival 21.1 vs 12.7 months), durvalumab + enfortumab vedotin before and after surgery in bladder cancer in patients who cannot receive cisplatin, sacituzumab tirumotecan in endometrial cancer, and chemotherapy-free pembrolizumab in dMMR endometrial cancer. Antibody–drug conjugates beat chemotherapy in five different cancers.

Will the ESMO 2026 results change my treatment right away?

Mostly not. Congress data are first analyses; the approval, guideline and reimbursement processes for drugs can take months or years. Some results come only from trials conducted in China and need to be confirmed. Discuss any change to your treatment with your oncologist.

Is it safe to receive immunotherapy less often?

The MOIO trial presented at ESMO 2026 is a warning on this point: in responding patients, spacing immunotherapy out to once every 12 weeks could not show equivalence for progression-free survival, and the rate of loss of life was higher, at 19.9% vs 10.5%. For this reason, the intervals should not be lengthened on one's own.

Are these new treatments available in Turkey?

Pembrolizumab, durvalumab, ribociclib, darolutamide and osimertinib are licensed in Turkey, but most of the new uses presented at ESMO 2026 are not yet approved or reimbursed. mRNA therapy, divarasib, ivonescimab and the new antibody–drug conjugates are not licensed; they can be obtained through clinical trials or by importing medicines from abroad.

Glossary

Hazard ratio (HR)

The ratio of the rates at which an event (progression, recurrence or loss of life) occurs in two groups; below 1 favors the new treatment.

Median survival

The time by which half of patients have lived longer and half have lived for a shorter time.

Progression-free / event-free survival

The time that passes without the disease progressing or coming back and without loss of life.

Antibody–drug conjugate

Chemotherapy attached to an antibody that binds to a target on the cancer cell; it carries the drug directly to the tumor cell.

Bispecific antibody

An antibody that can bind to two different targets at the same time; for example, tarlatamab links the tumor cell to an immune cell.

dMMR

A deficiency in the tumor's DNA repair system; these tumors respond very well to immunotherapy.

Pathological complete response

No living tumor cells remaining in the tissue removed after treatment given before surgery.

Interim analysis

A statistical assessment carried out at a pre-planned point before the trial is complete; results may change with longer follow-up.