Lung immunotherapy: before, after or both

A 63-year-old engineer, a former smoker, comes to the consultation with a CT scan showing a 5-centimeter mass in the right upper lobe. PET-CT shows uptake in a mediastinal lymph node on the same side, and endobronchial ultrasound-guided biopsy confirms squamous cell carcinoma with N2 disease. There are no distant metastases. The surgeon thinks resection is possible. The oncologist points out that today there are three paths with immunotherapy: treat before surgery, treat after, or treat at both times. The patient wants to know which is best, and whether he can simply have surgery right away. The scene is fictional and was written for teaching.

This text uses the scenario as a setting for clinical reasoning for students and health professionals. The question comes from Dr. Alexei's list for the oncology series: how to choose between a neoadjuvant, adjuvant or perioperative strategy in resectable lung cancer? The reference trials are CheckMate 816, with nivolumab before surgery, AEGEAN, with durvalumab before and after, and IMpower010, with atezolizumab after surgery and chemotherapy. The text does not replace your country's guideline or your institution's protocol. If you play the case of the day on NEXO, you will recognize the challenge: choosing the right order of steps is part of the reasoning, not a detail.

Key points

• In resectable non-small cell lung cancer, immunotherapy can come before surgery, after it, or at both times.
• CheckMate 816 showed that three cycles of nivolumab with chemotherapy before surgery increase pathological complete response and event-free survival.
• AEGEAN added durvalumab before and after surgery and also improved event-free survival and pathological response.
• IMpower010 showed a benefit of adjuvant atezolizumab after chemotherapy, especially with PD-L1 positive.
• Before any choice, EGFR, ALK and PD-L1 need to be available, because they change the strategy.
• The decision needs to be made by the team before surgery, because the neoadjuvant path no longer exists once the tumor is out.

Why this is a reasoning scenario

For a long time, the patient with resectable lung cancer followed a fixed sequence. Surgery came first and, depending on the report, chemotherapy came after. Chemotherapy before surgery existed, but the gain was similar to that of chemotherapy after, and many services preferred to operate right away. The arrival of immune checkpoint inhibitors changed this conversation, because the order of treatments began to make a difference in the outcome.

Today, the same patient may be a candidate for three paths with phase III evidence. This creates a reasoning problem that did not exist before. The choice needs to be made at the time of diagnosis, by a team that includes surgery, clinical oncology, radiology, pathology and pulmonology. If the patient has surgery right away, the neoadjuvant option disappears. If the patient receives treatment before, they need to reach surgery in good condition.

The reasoning here has three layers. The first is knowing what each trial tested and in which population. The second is recognizing the biomarkers that exclude immunotherapy or favor another strategy. The third is weighing practical risks, such as toxicity before surgery and the chance that the patient will not reach it. The post about the multidisciplinary team in lung cancer shows why this decision cannot belong to a single specialty.

The error the scenario wants to avoid

The most frequent error is not choosing the wrong regimen. It is letting the choice happen by omission. The patient is referred straight to the surgeon, has surgery in two weeks, and only afterward does someone ask whether they could have received treatment before. In a real-world study with data from about 280 clinics in the United States, neoadjuvant chemoimmunotherapy rose from 8.4 to 13.8 per 100 operated patients between 2022 and 2023, and total use stayed below 30 per 100 eligible patients Desai and colleagues, 2025.

Three strategies and what each one promises

Each strategy has a different biological and practical rationale. Understanding this rationale helps explain the choice to the patient and anticipate problems.

Neoadjuvant

Treating before surgery exposes the immune system to the whole tumor, with all antigens still present, which can produce a broader response. It also allows the response to be measured in the surgical specimen, through complete or major pathological response. The risk is progression during treatment or toxicity that delays or cancels surgery Ni and colleagues, 2021.

Adjuvant

Treating after surgery ensures the tumor is removed without delay and allows decisions to be made with complete pathological staging. The disadvantage is that the immune system no longer has the primary tumor as a target, and some patients do not reach treatment because of postoperative complications.

Perioperative

Treating before and after combines both rationales. Perioperative studies have shown improved event-free survival, but it is still debated how much the adjuvant phase adds for those who already had a complete response in the specimen Peng and colleagues, 2023.

• Neoadjuvant: response measured in the specimen, short treatment, risk of not reaching surgery.
• Adjuvant: surgery without delay, decision with pathological stage, risk of not starting afterwards.
• Perioperative: both phases, longer treatment, more time exposed to toxicity.
• In all of them: biomarkers defined beforehand and the team deciding together.

CheckMate 816: treating before surgery

CheckMate 816 was an open-label phase III study. Patients with stage IB to IIIA were randomized to receive nivolumab 360 mg with platinum doublet chemotherapy, or chemotherapy alone, every 3 weeks for three cycles. Surgery was to take place up to 6 weeks after the end of neoadjuvant treatment (Mitsudomi and colleagues, 2024). The study included 358 patients, about 63 in every 100 with stage IIIA (Mayenga and colleagues, 2024).

Complete pathological response, defined as the absence of viable tumor cells in the tumor and lymph nodes, was 24 in every 100 patients with nivolumab and 2.2 in every 100 with chemotherapy alone. Median event-free survival was 31.6 months versus 20.8 months, with a hazard ratio of 0.63 (Mayenga and colleagues, 2024). Surgery was performed in about 83 in every 100 patients in the nivolumab group and in about 75 in every 100 in the control group, and grade 3 or 4 adverse events were similar in both groups (Mayenga and colleagues, 2024).

PD-L1 and response

Complete pathological response was higher with PD-L1 of 1 or more in every 100 cells, about 33 in every 100 patients, but it also appeared with negative PD-L1, about 17 in every 100 (Mayenga and colleagues, 2024). This finding helps explain why negative PD-L1 does not rule out the neoadjuvant strategy, although the gain is smaller.

Smaller studies help show what comes next. In NADIM, a Spanish phase II study with nivolumab and chemotherapy in stage IIIA, overall survival at 36 months was about 82 in every 100 patients. Undetectable circulating tumor DNA after neoadjuvant treatment was associated with better survival, performing better than radiological response (Provencio and colleagues, 2022). In the first study with neoadjuvant nivolumab alone, 8 of the 9 patients with major pathological response were alive and disease-free at 5 years (Rosner and colleagues, 2023).

AEGEAN: treating before and after

AEGEAN tested durvalumab 1500 mg or placebo with chemotherapy every 3 weeks for four cycles before surgery. After surgery, patients continued with durvalumab or placebo every 4 weeks for up to 12 cycles (Mitsudomi and colleagues, 2026). The primary analysis excluded patients with documented EGFR or ALK alterations, who had been enrolled before a protocol amendment (Mitsudomi and colleagues, 2026).

Perioperative durvalumab improved event-free survival, with a hazard ratio of 0.68, and pathologic complete response. In the final analysis of the global population, pathologic complete response was 17.2 in 100 with durvalumab versus 4.3 in 100 with placebo (Mitsudomi and colleagues, 2026). Immune-mediated pneumonitis of any grade was reported in about 5 in 100 patients during the adjuvant period in the Japanese subgroup (Mitsudomi and colleagues, 2026).

What the adjuvant phase adds

The perioperative design does not separate the effect of each phase. The patient who received durvalumab before also received it afterward. So the question of who actually needs the adjuvant phase remains open, and reviews on the topic point to this doubt as one of the main ones in the field (Peng and colleagues, 2023). The post about perioperative pembrolizumab in KEYNOTE-671 discusses the same question in another study.

Study	Strategy and dose	Pathologic complete response	Primary outcome
CheckMate 816	Nivolumab 360 mg with chemotherapy, 3 cycles before	24 versus 2.2 in 100	Event-free survival, hazard ratio 0.63
AEGEAN	Durvalumab 1500 mg with chemotherapy, 4 cycles before and up to 12 after	17.2 versus 4.3 in 100	Event-free survival, hazard ratio 0.68
IMpower010	Atezolizumab 1200 mg every 3 weeks, up to 16 cycles after chemotherapy	not applicable	Disease-free survival, hazard ratio 0.66 in stage II to IIIA with positive PD-L1

IMpower010: treating after surgery

IMpower010 included operated patients with stage IB to IIIA who received one to four cycles of adjuvant chemotherapy. After that, they were randomized to atezolizumab 1200 mg every 3 weeks for up to 16 cycles, or supportive care (Okada and colleagues, 2026).

Atezolizumab improved disease-free survival in stage II to IIIA with PD-L1 of 1 or more in 100 tumor cells, with a hazard ratio of 0.66. The benefit was greater with PD-L1 of 50 or more, with a hazard ratio of 0.43. In the intention-to-treat population, which included stage IB, the result did not reach significance (Okada and colleagues, 2026). These data led to the approval of adjuvant atezolizumab for stage II to IIIA with positive PD-L1, excluding EGFR and ALK in the European Union (Okada and colleagues, 2026).

The PD-L1 test matters

Different PD-L1 tests can raise doubt. In IMpower010, the SP263 and 22C3 tests agreed in 83 in 100 samples at the threshold of 1 and in 92 in 100 at the threshold of 50, with a similar benefit from atezolizumab by both methods (Zhou and colleagues, 2023). In practice, the most important thing is that the result exists before the decision.

One limitation of the adjuvant strategy is that it depends on the patient recovering well from surgery, completing chemotherapy and only then starting immunotherapy. Each step is a chance to lose the patient, a subject covered in the post about barriers to curative treatment.

Biomarkers that change the choice

Before talking about immunotherapy, the team needs to know whether the tumor has a targetable alteration. An expert consensus recommends high-throughput sequencing (NGS) before treatment and caution with neoadjuvant immunotherapy when there is a sensitizing EGFR mutation or ALK fusion. The same text recalls that, in Asia, 40 to 50 in 100 lung adenocarcinomas have an activating EGFR mutation (Qiu and colleagues, 2021).

For these patients, the choice changes. Adjuvant osimertinib has a survival benefit in tumors with mutated EGFR, as discussed in the post about the ADAURA study, and adjuvant alectinib has a role in ALK-positive tumors, the subject of the post about the ALINA study.

• EGFR: if mutated, immunotherapy loses its place and targeted treatment gains.
• ALK: if rearranged, the same reasoning.
• PD-L1: guides the size of the benefit and defines the adjuvant indication for atezolizumab.
• Histology: squamous or non-squamous changes the chemotherapy regimen.
• Clinical stage: defines eligibility for each study.

Does negative PD-L1 rule out immunotherapy?

A meta-analysis of neoadjuvant, adjuvant and perioperative studies showed improved pathological response and event-free survival at all PD-L1 levels. The hazard ratio for event-free survival was 0.80 with PD-L1 below 1 and 0.53 with PD-L1 of 1 or more. The overall survival benefit was significant with PD-L1 positive and uncertain with PD-L1 negative (Zhang and colleagues, 2025). PD-L1 negative, therefore, reduces the expected gain, but does not end the discussion.

Risks of immunotherapy before surgery

Neoadjuvant immunotherapy can cause serious immune-mediated adverse events. Chinese recommendations on the topic note that these events can lead to delay or cancellation of surgery, additional disease and, rarely, death (Ni and colleagues, 2021). Pneumonitis, hepatitis, colitis, thyroiditis and adrenal insufficiency are the events of most concern before thoracic surgery.

In CheckMate 816, however, the combination of nivolumab and chemotherapy was no more toxic than chemotherapy alone, and the proportion of patients who underwent surgery was even higher in the nivolumab group (Mayenga and colleagues, 2024). The risk exists, but does not by itself justify abandoning the neoadjuvant strategy.

What to monitor during treatment

• Thyroid function, blood glucose, liver enzymes and cortisol before and during the cycles.
• New respiratory symptoms, which may be pneumonitis and not progression.
• Diarrhea, which may indicate immune-mediated colitis.
• Restaging before surgery, to detect progression.
• Communication with the surgeon about any event that could change the date.

The post about diagnostic pause helps remind us that a new symptom during immunotherapy has more than one possible explanation.

How to choose in practice

The studies did not compare the three strategies with each other. Every comparison between neoadjuvant, adjuvant and perioperative is indirect. Even so, some principles guide the choice.

For the patient with bulky disease or N2, like the one in the opening case, the neoadjuvant or perioperative strategy has strong appeal: it offers a chance of tumor shrinkage, measures the response and treats micrometastatic disease early. For the patient already operated on without prior treatment, the adjuvant strategy with chemotherapy followed by atezolizumab is the available path when PD-L1 is positive and there is no EGFR or ALK. For small tumors without lymph node involvement, immunotherapy has a smaller role.

Back to the engineer in the opening case

The patient has squamous cell carcinoma, a 5-centimeter tumor and N2 disease confirmed by needle biopsy. He meets the eligibility criteria for CheckMate 816 and AEGEAN. Before choosing, the team checks for EGFR or ALK, which is rare in squamous cell carcinoma but not impossible, and requests PD-L1. With these results, the conversation with him can be honest: treating before surgery greatly increases the chance that no viable tumor remains in the specimen and reduces the risk of recurrence, at the cost of postponing the operation by about two to three months and exposing the patient to immune-mediated effects.

His question, whether he can simply have surgery right away, also deserves a clear answer. He can, and surgery will remain the center of treatment. But operating first closes the door on the neoadjuvant strategy, and adjuvant immunotherapy will depend on PD-L1 and on recovery after surgery. Explaining this trade-off, without pressure, is part of care.

Questions for the team meeting

• Was mediastinal staging confirmed by tissue?
• Are EGFR, ALK and PD-L1 available?
• Is the tumor resectable now, and will it remain so if there is a delay?
• Does the patient have active autoimmune disease or another contraindication to immunotherapy?
• What is the service's capacity for follow-up during the cycles?
• Does the patient understand the options and the risks of each one?

In the real-world study, 18-month distant metastasis-free survival was similar between the neoadjuvant and adjuvant groups, about 80 and 83 per 100, and PD-L1 testing was done less often in the neoadjuvant group (Desai et al., 2025). This finding does not define which strategy is better, but it shows that biomarker testing still fails precisely where it should come first. The post on clinical decision-making deals with this kind of choice made with incomplete information.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code. The choice between neoadjuvant, adjuvant and perioperative is a good exercise because it depends on clues that appear at different moments: clinical stage, biomarkers, patient preference and the pathology report. The post on differential diagnosis shows how to train with progressive clues.

Three exercises with cases

Change the biomarker. Say what changes if the panel shows an EGFR exon 19 deletion.

Change the timing. Say what you would do if the patient had already been operated on without prior treatment and the pathology report showed N2.

Change the course. Say how to act if the patient develops grade 2 pneumonitis after the second neoadjuvant cycle.

The post on diagnostic uncertainty helps you discuss options without direct comparison. The limits of use of the game are on the NEXO Economy page. If you research the teaching of clinical reasoning, the page for Researchers explains how to get in touch.

Frequently asked questions

What is the dose of nivolumab in CheckMate 816?

Nivolumab 360 mg with platinum doublet chemotherapy every 3 weeks, for three cycles before surgery (Mitsudomi et al., 2024). Check the package insert in your country before prescribing.

Is the perioperative strategy better than the neoadjuvant one?

There is no direct comparison. The perioperative studies do not separate the effect of the adjuvant phase, and this question remains open (Peng et al., 2023).

Should patients with mutated EGFR receive immunotherapy before surgery?

In general, no. Consensus guidelines recommend caution, and AEGEAN excluded these patients from the main analysis (Qiu et al., 2021) (Mitsudomi et al., 2026).

When should adjuvant atezolizumab be used?

After surgery and adjuvant chemotherapy, in stage II to IIIA with positive PD-L1, for up to 16 cycles (Okada et al., 2026).

Does a complete pathological response guarantee a cure?

It does not guarantee it, but it is associated with a better prognosis. Smaller studies show that a greater response and undetectable tumor DNA after treatment indicate a better course (Rosner et al., 2023) (Provencio et al., 2022).

Final checklist

• I confirmed the stage, including the mediastinum, with tissue when necessary.
• I requested EGFR, ALK and PD-L1 before the decision.
• I took the case to the meeting before scheduling surgery.
• I discussed the three strategies with the patient.
• I reviewed contraindications to immunotherapy.
• I planned toxicity monitoring tests during the cycles.
• I agreed with the surgeon on the time frame after the last cycle.
• I defined who decides on the adjuvant phase after the pathology report.

Want to practice reasoning with progressive clues? Play the case of the day.

References

1. Desai A, Schwed K, Kalesinskas L, et al. Clinical Outcomes of Perioperative Immunotherapy in Resectable Non-Small Cell Lung Cancer. JAMA Netw Open. 2025. PMID 40587130.
2. Mayenga M, Pedroso AR, Ferreira M, et al. The CheckMate 816 trial: a milestone in neoadjuvant chemoimmunotherapy of nonsmall cell lung cancer. Breathe (Sheff). 2024. PMID 39534491.
3. Mitsudomi T, Heymach JV, Horio Y, et al. Perioperative Durvalumab for Resectable NSCLC: Updated Outcomes for Japanese Patients in AEGEAN. Cancer Sci. 2026. PMID 42802029.
4. Mitsudomi T, Ito H, Okada M, et al. Neoadjuvant nivolumab plus chemotherapy in resectable non-small-cell lung cancer in Japanese patients from CheckMate 816. Cancer Sci. 2024. PMID 38098261.
5. Ni J, Huang M, Zhang L, et al. Clinical recommendations for perioperative immunotherapy-induced adverse events in patients with non-small cell lung cancer. Thorac Cancer. 2021. PMID 33787090.
6. Okada M, Sugawara S, Watanabe Y, et al. IMpower010: 5-Year Outcomes of Atezolizumab in Japanese Patients With Resected Stage IB-IIIA Non-Small Cell Lung Cancer. Cancer Sci. 2026. PMID 41416818.
7. Peng Y, Li Z, Fu Y, et al. Progress and perspectives of perioperative immunotherapy in non-small cell lung cancer. Front Oncol. 2023. PMID 36761954.
8. Provencio M, Serna-Blasco R, Nadal E, et al. Overall Survival and Biomarker Analysis of Neoadjuvant Nivolumab Plus Chemotherapy in Operable Stage IIIA Non-Small-Cell Lung Cancer (NADIM phase II trial). J Clin Oncol. 2022. PMID 35576508.
9. Qiu B, Cai K, Chen C, et al. Expert consensus on perioperative immunotherapy for local advanced non-small cell lung cancer. Transl Lung Cancer Res. 2021. PMID 34733623.
10. Rosner S, Reuss JE, Zahurak M, et al. Five-Year Clinical Outcomes after Neoadjuvant Nivolumab in Resectable Non-Small Cell Lung Cancer. Clin Cancer Res. 2023. PMID 36794455.
11. Zhang Z, Lin Y, Chen S. Efficacy of neoadjuvant, adjuvant, and perioperative immunotherapy in non-small cell lung cancer across different PD-L1 expression levels: a systematic review and meta-analysis. Front Immunol. 2025. PMID 40463367.
12. Zhou C, Srivastava MK, Xu H, et al. Comparison of SP263 and 22C3 immunohistochemistry PD-L1 assays for clinical efficacy of adjuvant atezolizumab in non-small cell lung cancer: results from the randomized phase III IMpower010 trial. J Immunother Cancer. 2023. PMID 37903590.