SWOG S1801 is a randomized clinical trial that changed the standard of care for advanced melanoma by showing that giving the immunotherapy drug pembrolizumab before and after surgery is substantially better than giving it only after surgery. At a median follow-up of about 15 months, patients who received neoadjuvant (pre-surgery) plus adjuvant (post-surgery) pembrolizumab had roughly a 42% lower rate of their cancer returning or progressing compared with patients treated after surgery alone. The trial’s results, published in the New England Journal of Medicine, have reshaped how oncologists think about timing immunotherapy in melanoma and sparked a broader rethinking of surgical sequencing across other tumor types.
What the Trial Tested and What It Found
SWOG S1801 enrolled 313 patients with resectable stage III or IV melanoma at centers across the United States. All patients were going to receive pembrolizumab, a checkpoint inhibitor that helps the immune system recognize and attack cancer cells. The question was when to give it relative to surgery. One group (154 patients) received three doses of pembrolizumab before surgery, then continued with additional doses afterward. The other group (159 patients) went straight to surgery and received all their pembrolizumab afterward. The total amount of drug was the same in both arms; only the sequence differed.
The difference in outcomes was striking. At two years, about 72% of patients in the neoadjuvant-adjuvant group were free of disease events (recurrence, progression, or death) compared with roughly 49% in the adjuvant-only group. The hazard ratio was 0.58, meaning the risk of an event was cut nearly in half by giving some of the immunotherapy before the surgeon operated.1PubMed Central. Neoadjuvant-Adjuvant or Adjuvant-Only Pembrolizumab in Advanced Melanoma The trial was stopped early at its first interim analysis because the benefit had already crossed the pre-specified statistical threshold.
Importantly, giving pembrolizumab before surgery did not seem to add toxicity. The rate of serious (grade 3 or higher) treatment-related side effects was actually slightly lower in the neoadjuvant-adjuvant group, at about 12%, compared with 14% in the adjuvant-only group.2N Engl J Med. Neoadjuvant-Adjuvant or Adjuvant-Only Pembrolizumab in Advanced Melanoma Surgery was not delayed or complicated in clinically meaningful ways. This addressed one of the biggest concerns clinicians had going in: that immune-related side effects from neoadjuvant treatment might make surgery riskier or force patients to miss their surgical window.
Why Immunotherapy Before Surgery May Work Better
The fact that the same drug worked better simply by shifting its timing raises a natural question: what is different about giving immunotherapy while the tumor is still in the body? The leading explanation centers on the immune system’s ability to learn. When the primary tumor is still present, it provides an enormous library of protein fragments, called antigens, that immune cells can train against. Checkpoint inhibitors like pembrolizumab essentially remove the brakes that cancer puts on T cells. If those T cells are activated while they have full access to the tumor’s diverse antigen supply, they can mount a broader, more varied attack.
Research comparing patients treated before versus after surgery has found that those who received neoadjuvant immunotherapy had a greater expansion of T-cell clones in their blood. This broader immune repertoire is thought to be important because melanoma cells are genetically heterogeneous: some cells carry mutations that others do not. An immune response trained against the whole tumor is more likely to catch metastatic cells hiding elsewhere in the body that carry different mutations than the bulk of the primary mass.3AACR Journals. The Promise of Neoadjuvant Immunotherapy and Surgery for Cancer Treatment Once surgery removes the main tumor, those primed T cells continue circulating and surveilling for residual cancer. In the adjuvant-only approach, the tumor is removed first, which may eliminate the very training ground the immune system needs.
Reading the Tumor After Treatment
One practical advantage of the neoadjuvant approach is that surgeons and pathologists get a live readout of whether the drug is working. After the pre-surgery doses, the tumor is removed and examined under a microscope. If the treatment devastated the cancer, pathologists see very few or no remaining viable tumor cells. If it did not work, the tumor looks largely untouched. This information, called the pathologic response, has proven to be a powerful predictor of long-term outcomes.
Studies of neoadjuvant immunotherapy in melanoma have classified responses by how much live tumor remains. A major pathologic response (10% or fewer viable tumor cells) or a complete pathologic response (no viable tumor at all) signals a strong immune attack. In follow-up data from a neoadjuvant pembrolizumab cohort, patients who achieved 10% or fewer viable tumor cells had a five-year disease-free survival rate of about 75%.4PubMed Central. Long-term outcomes to neoadjuvant pembrolizumab based on pathological response for patients with resectable stage III/IV cutaneous melanoma Radiological imaging before surgery (CT scans, for instance) has also shown correlation with the pathologic findings: tumors that shrank on imaging tended to show more destruction under the microscope.5Annals of Oncology. Pathological response and tumour bed histopathological features correlate with survival following neoadjuvant immunotherapy in stage III melanoma
This ability to see how the cancer responded creates a feedback loop that the adjuvant-only model simply cannot offer. If you give immunotherapy only after surgery, the tumor is already gone. There is no specimen to evaluate. Clinicians are left guessing whether the patient’s immune system actually responded to the drug, or whether the cancer is silently regrouping.
Tailoring Surgery and Followup Based on Response
Some researchers have taken the pathologic response concept a step further, asking whether patients with excellent responses can be spared more aggressive surgery or skip adjuvant therapy altogether. The PRADO trial tested exactly this idea. Patients with high-risk stage III melanoma received neoadjuvant combination immunotherapy (ipilimumab plus nivolumab, a different checkpoint combination). Those who achieved a major pathologic response in their largest lymph node metastasis had their full lymph node dissection and adjuvant therapy omitted.6Nature Medicine. Personalized response-directed surgery and adjuvant therapy after neoadjuvant ipilimumab and nivolumab in high-risk stage III melanoma: the PRADO trial The idea is that if the immune system has already obliterated the cancer, extensive surgery and months of additional drug therapy may cause unnecessary harm without adding benefit.
This response-adapted strategy represents a significant departure from the one-size-fits-all approach historically used in melanoma. SWOG S1801 established that neoadjuvant therapy improves outcomes on average. PRADO refined the idea by using the response data to personalize what happens next. The two trials are complementary: one shows the benefit of the neoadjuvant approach across the board, and the other shows how that approach can be used to de-escalate care in the best responders.
Circulating Tumor DNA as a Monitoring Tool
Beyond what pathologists can see in the removed tissue, researchers have been exploring blood-based markers to track how well neoadjuvant immunotherapy is working. Circulating tumor DNA (ctDNA), fragments of cancer-derived DNA floating in the bloodstream, has emerged as a particularly promising tool. A study of patients receiving neoadjuvant immunotherapy for stage III melanoma found that ctDNA dynamics were remarkably accurate at predicting who would and would not relapse.
Among patients who had detectable ctDNA before treatment, 89% saw their ctDNA levels drop to undetectable by six weeks after surgery. None of those patients experienced recurrence during follow-up. The two patients whose ctDNA remained detectable after surgery both relapsed within eight months. When pre-treatment ctDNA was detectable, the test’s sensitivity and specificity for predicting recurrence were both perfect in this cohort.7PubMed Central. Circulating tumour DNA dynamics predict recurrence in stage III melanoma patients receiving neoadjuvant immunotherapy The numbers come from a small group of 19 patients, so they need validation in larger studies, but the pattern is striking enough to warrant continued investigation. If this holds up, ctDNA monitoring could serve as an early warning system, flagging patients who need intensified treatment well before a visible recurrence appears on a scan.
Quality of Life After Treatment
Clinical trials usually focus on survival and recurrence, but for the person undergoing treatment, daily functioning and well-being matter enormously. A cross-sectional study comparing long-term quality of life between patients treated with neoadjuvant versus adjuvant immunotherapy found notable differences favoring the neoadjuvant approach. Patients who had received neoadjuvant treatment reported better physical, role, cognitive, and social functioning, along with significantly less fatigue. The fatigue difference held up even after adjusting for other factors that could influence quality of life.8European Journal of Cancer. Cross-sectional study of long-term Health-Related Quality of Life in stage III melanoma patients receiving neo-adjuvant versus adjuvant immune checkpoint inhibitors
One possible explanation is that patients who responded well to neoadjuvant therapy may have had less extensive surgery. Removing fewer lymph nodes, for instance, can reduce the risk of long-term side effects like lymphedema (chronic swelling). Another factor may be psychological: knowing that the cancer showed a strong response before surgery could reduce the anxiety and uncertainty that color the recovery period. The study’s sample was modest, so these results should be interpreted cautiously, but they reinforce a pattern that clinicians are seeing across multiple cancer types where neoadjuvant strategies allow de-escalation of subsequent care.
The Statistical Debate
No landmark trial escapes scrutiny, and SWOG S1801 is no exception. A critique published in Translational Oncology raised concerns about potential time biases in the trial’s analysis. The core issue: in the neoadjuvant-adjuvant arm, patients received about nine weeks of pembrolizumab before surgery, whereas patients in the adjuvant-only arm went straight to surgery and began pembrolizumab afterward. Because event-free survival was measured from the time of randomization (rather than from surgery), the neoadjuvant group effectively had a built-in head start. Any events that might have occurred during that pre-surgery treatment period were counted differently across the two arms.9PubMed Central. Neoadjuvant followed by adjuvant pembrolizumab in melanoma: time biases in the data analysis of the SWOG S1801 trial
This type of methodological question comes up frequently in neoadjuvant trial design and does not necessarily invalidate the findings. The trial investigators chose randomization as the starting point precisely because it avoids other biases (for example, patients who progress before surgery and never get operated on would be excluded from a surgery-based landmark, which could artificially inflate results for the neoadjuvant arm). The critique is a reminder that the magnitude of the benefit may be somewhat sensitive to the analytical framework chosen, but the direction of the effect, that neoadjuvant therapy helps, has been consistent across analyses and is supported by converging data from other trials.
How This Fits Into the Broader Melanoma Landscape
SWOG S1801 was not the only trial to test neoadjuvant immunotherapy in melanoma, though it was the first large randomized trial to definitively show an event-free survival advantage. The NADINA trial, using a different checkpoint inhibitor combination (ipilimumab plus nivolumab before surgery compared to nivolumab alone after surgery), also demonstrated improved event-free survival with the neoadjuvant approach.10SpringerLink / Annals of Surgical Oncology. Implementing Paradigm Shifting Clinical Trials: Real-World Outcomes of Neoadjuvant Versus Adjuvant Immune Checkpoint Inhibitors in Advanced Melanoma Patients Real-world retrospective studies are now beginning to confirm that the benefits seen in these controlled trials translate to everyday clinical practice, where patients are more heterogeneous and treatment adherence is more variable than in a trial setting.
The convergence of evidence from S1801, NADINA, PRADO, and other studies has been enough to shift treatment guidelines. Many melanoma centers now offer neoadjuvant immunotherapy as a standard option for patients with resectable advanced disease. The conversation between surgeon and medical oncologist has changed: instead of “we’ll cut it out and then treat,” it is increasingly “let’s treat first, see how it responds, then operate accordingly.” For patients, this means the surgical plan may actually be informed by how their specific cancer behaves in real time.
Standardizing Pathologic Response Scoring
As neoadjuvant immunotherapy becomes standard practice, one practical challenge has been ensuring that pathologists everywhere are evaluating response in the same way. After a patient receives neoadjuvant therapy and undergoes surgery, the removed tissue needs to be sliced, stained, and examined to determine how much viable tumor remains. Different institutions historically used different scoring systems, making cross-trial comparisons difficult and raising concerns about consistency in everyday clinical use.
A joint effort by the Society for Immunotherapy of Cancer (SITC) and the International Neoadjuvant Melanoma Consortium (INMC) has worked to update and harmonize pathologic response scoring into a standardized system that can be applied across tumor types.11PubMed Central. Updated pan-tumor guidelines for neoadjuvant scoring of pathologic response: a joint SITC and INMC effort In parallel, researchers have investigated whether pathologists really need to examine every single tissue slide to get an accurate response score. A study focusing on melanoma lymph node specimens found that a practical protocol works well: lymph nodes under 3 cm should be examined in their entirety, while larger nodes can be adequately assessed with a representative cross-sectional slice, capping the total number of slides at about 20. This refined approach maintained high accuracy while significantly reducing the workload.12PubMed. Pathological response calculation assessment remains accurate with reduced tumor bed examination after neoadjuvant immunotherapy in clinically detectable stage III melanoma
These efforts matter because pathologic response is no longer just a research endpoint. It is becoming a clinical decision tool. If a patient achieves a major pathologic response, they may be spared additional surgery or months of adjuvant therapy, as PRADO demonstrated. For that decision to be reliable, the scoring needs to be reproducible whether the patient is treated at a large academic cancer center or a community hospital. Getting pathology protocols right is part of making neoadjuvant immunotherapy scalable.
What SWOG S1801 Did Not Answer
For all its impact, the trial left several questions open. The initial publication reported event-free survival at a relatively early time point; longer follow-up is needed to confirm whether the advantage translates to an overall survival benefit. Event-free survival is meaningful to patients because it measures how long they stay cancer-free, but it is not the same as living longer overall. Some patients who recur can be treated again successfully, so a difference in recurrence timing does not always map cleanly onto a difference in total lifespan. Updated survival data from S1801 are awaited.
The trial also used pembrolizumab as a single agent. Other trials, like NADINA, have tested combination checkpoint blockade (ipilimumab plus nivolumab), which is more potent but comes with significantly more side effects. It remains unclear which neoadjuvant regimen is optimal for which patients. A fit, younger patient might tolerate combination therapy and gain a higher pathologic response rate. An older patient with autoimmune conditions might do better with single-agent pembrolizumab and its milder toxicity profile. Head-to-head comparisons of different neoadjuvant regimens are still needed.
There is also the question of who benefits most. SWOG S1801 enrolled patients with stage III and IV resectable melanoma, but within that group, there may be subsets where the neoadjuvant advantage is larger or smaller. Biomarker studies, including work with ctDNA and tumor-infiltrating lymphocyte profiles, are ongoing to identify patients who are most and least likely to respond. Until those predictive tools mature, the broad recommendation applies to the whole eligible population, even though the actual benefit for any individual patient is uncertain.