Not All Immune Checkpoint Inhibitor Side Effects Are Equal for Survival

When a cancer patient develops a side effect during immunotherapy, oncologists often see it as a good sign — evidence that the drug has successfully unleashed the immune system against the tumor. But a new study suggests this popular assumption is only half the story. Not all immune-related adverse events are created equal, and some may actually signal worse survival rather than better.

Immune checkpoint inhibitors, or ICIs, have transformed the treatment of advanced cancer over the past decade. These drugs, which include antibodies targeting PD-1 and PD-L1, work by releasing the molecular brakes that tumors place on T cells. But the same immune activation that attacks cancer can also strike healthy tissue, producing immune-related adverse events, or irAEs. According to the literature, these occur in roughly 70 percent of patients receiving anti-PD-1 or PD-L1 therapy, with thyroid inflammation and skin toxicities among the most common.

For years, researchers have debated what these side effects mean for prognosis. Some studies report a positive association between irAEs and survival, while others find that certain irAEs are linked to worse outcomes and treatment interruption. The picture has remained murky, partly because most analyses lump all adverse events together rather than examining each one individually.

Now, a team of medical oncologists at Saglik Bilimleri University’s Prof. Dr. Cemil Tascioglu City Hospital in Istanbul has taken a more granular approach. In a retrospective study published in BMC Cancer, Mert Erciyestepe and colleagues analyzed data from 300 patients with metastatic malignancy treated between January 2020 and January 2024. Their goal was to dissect the relationship between individual irAEs and survival in patients receiving ICI treatment in first- and second-line settings.

The headline finding is striking. Patients who experienced any adverse event had a median overall survival of 25.0 months, compared with just 9.0 months in patients who developed no adverse events at all. The difference was highly statistically significant, with a p-value below 0.001. On the surface, this supports the conventional wisdom that side effects correlate with better outcomes.

But when the researchers broke the data down by specific adverse event, the story became far more nuanced. Patients who developed hypothyroidism had a median overall survival of 41.0 months, compared with 23.0 months in those who did not — a dramatic difference. Similarly, patients who experienced skin toxicity had a median overall survival of 48.0 months, the longest of any group examined. Both findings were statistically significant.

Not every adverse event was a good omen, however. Patients who developed elevated AST/ALT levels — a sign of liver inflammation — had significantly shorter survival, with a p-value of 0.015. Even more concerning, patients who developed pneumonitis, an inflammation of the lung tissue that is one of the most feared complications of checkpoint inhibitor therapy, had significantly shorter survival, with a p-value below 0.001. Meanwhile, the presence of mucositis, acute kidney injury, arthritis, adrenal insufficiency, and hypophysitis did not show any statistically significant difference in overall survival.

The researchers used standard survival analysis techniques to reach these conclusions. They employed the Kaplan-Meier method and log-rank tests to compare overall survival and progression-free survival between clinical groups, used independent t-tests for pairwise comparisons, and applied multivariate Cox regression to assess the risk of death across various clinical variables. This combination of approaches helps control for confounding factors that might otherwise distort the apparent relationship between adverse events and survival.

What might explain why some irAEs correlate with better survival while others correlate with worse? One possibility is that certain adverse events, like thyroid dysfunction and skin toxicity, reflect a broadly activated immune system that is also attacking the tumor. Hypothyroidism in particular is easily managed with hormone replacement, so it poses little threat to the patient while potentially serving as a biomarker of immune activation. In contrast, pneumonitis and hepatitis may reflect a dangerous loss of immune self-tolerance that forces treatment interruption, denying the patient the full benefit of therapy while simultaneously threatening vital organs.

The clinical implications are significant. If individual irAEs can serve as indicators of treatment response and survival, oncologists may be able to use them as real-time biomarkers, adjusting monitoring intensity and treatment strategy based on which side effects appear. A patient who develops hypothyroidism or a mild skin rash might be reassured that their immune system is engaged, while a patient who develops pneumonitis might warrant closer surveillance and more aggressive management. The authors suggest that survival can potentially be predicted by monitoring irAEs, though they caution that the number of studies on this subject remains relatively small.

Limitations of the study are worth noting. As a retrospective, single-center analysis of 300 patients, the findings may not generalize to all populations or treatment settings. The authors themselves call for larger, multicenter studies to confirm and extend their results. Still, the study adds an important layer of granularity to a controversial topic, moving the field beyond the simplistic question of whether adverse events are good or bad for survival. The answer, it turns out, depends on which adverse event — and that distinction could ultimately help clinicians make better decisions for their patients.

Subject of Research: Distribution of adverse effects of immune checkpoint inhibitors and their impact on survival: some with better but some with worse survival

Article Title: Distribution of adverse effects of immune checkpoint inhibitors and their impact on survival: some with better but some with worse survival

Article References: Erciyestepe, M., Afsar Satis, N., Ozturk, A. E., Dinc Sonusen, S., Aydin, O., & Erturk, K. (2026). Distribution of adverse effects of immune checkpoint inhibitors and their impact on survival: some with better but some with worse survival. BMC Cancer. https://doi.org/10.1186/s12885-026-16954-8

Image Credits: AI Generated

DOI: 10.1186/s12885-026-16954-8

Keywords: Distribution, adverse, effects, immune, checkpoint, inhibitors, impact, survival, some, better, worse, scientific research