Immune checkpoint inhibitors (ICIs) can restore the immune response against cancer by blocking inhibitory molecules, such as cytotoxic T lymphocyte-associated protein 4 (CTLA-4), programmed death-1 (PD-1) and its ligand (PD-L1), expressed on immune and/or tumor cells, the so-called immune checkpoints (1–3). Their use has revolutionized the standard of care of cancer patients, providing therapeutic options for many advanced stage tumors considered otherwise untreatable (4–9). ICI treatment has been associated with excellent response rates and improved survival when administered as either first-line therapy or after other treatments (10–12). Some tumor characteristics, such as the expression of PD-L1 on cancer cells, a T-cell inflamed profile (T-cell infiltration), and the mutational and/or neoantigen burden are currently known to predict response to ICIs, although the identification of predictive biomarker for ICI-based therapy is still challenging (13, 14). In particular, the impact of patient-related factors, like sex, age or BMI, remains to be elucidated, as well as the predictive and/or prognostic role of ICI adverse events, to be considered “on-target” side effects (15).

Indeed, as indications for ICI therapy have expanded and numbers of treated patients have increased, a unique profile of toxicity has emerged, characterized by the occurrence of immune-mediated damage of several tissues and organs (16). These immune-related adverse events (irAEs), autoimmune in etiology, are reported in up to 50% and more of treated patients and can potentially affect all organs. Dermatologic, gastrointestinal, hepatic, and endocrine manifestations are the most frequently reported irAEs, while neurological, cardiac or pulmonary side effects rarely occur (16). In particular, thyroid disorders are among the most common endocrine irAEs, mostly under anti-PD-1/PD-L1 treatment, and include hyperthyroidism, hypothyroidism and destructive thyroiditis (thyrotoxicosis progressing to hypothyroidism) (17–19). Other less common endocrinopathies include hypophysitis, adrenal insufficiency, type 1 diabetes, and hypoparathyroidism (19–21). Although ICI-based therapies are typically well tolerated, the risk of potentially severe irAEs, compromising organ function and/or quality of life, is not negligible and increases with combination regimens (17).

Moreover, some (22, 23) but not all (24) studies have found an association between female sex and occurrence of irAEs.

A growing body of evidence suggests that overweight/obesity may be associated with increased immunotoxicity on the one hand (25) and with improved efficacy of immunotherapy on the other hand (26, 27). The mechanisms behind this unexpected favorable association, the so-called “obesity paradox”, are still not clear (28, 29), and the real impact of overweight on irAE development and efficacy still remains to be further defined. The present study was aimed at evaluating the occurrence of irAEs among cancer patients on ICI therapy according to baseline BMI and gender. Further, we analyzed survival outcome difference in these subgroups of patients.

The present study investigated the impact of gender and BMI on irAE development and efficacy in a cohort of patients with different types of cancer. In our single center cohort, 51 patients (39.2%) developed an irAE, thyroid dysfunction being the most common one, consistent with previously published real-world data (33–36).

With regard to gender differences, overall, irAEs were more frequently observed in female patients, with the relevant exception of thyroid disorders that occurred equally in both genders. Unlike other studies including a large number of patients with lung cancer (37–39), we found a female prevalence in irAEs, as reported in other series (23). Moreover, female patients experienced an earlier onset of irAEs than males.

Regarding the impact of BMI, we found that it was strongly associated with the occurrence of irAEs. Indeed, the prevalence of irAEs was significantly higher among overweight/obese patients compared to normal weight patients in both sexes, and patients with higher BMI were at increased risk of developing an irAE, with an OR of 3.182 compared to normal weight/underweight patients.

Evidence regarding the association between BMI and irAEs among cancer patients receiving ICIs is limited, and sometimes conflicting. Some studies reported a significantly higher incidence of irAEs in patients with higher BMI (25, 35, 39), while other studies failed to demonstrate such an association. Two meta-analyses explored the association between BMI and irAEs among patients with cancer receiving ICIs, and both concluded that high BMI was associated with a higher rate of irAEs (40, 41). Our study provides further evidence in support of a positive correlation between BMI and development of irAEs, reporting a 3-fold increase of the risk of irAEs among overweight/obese patients. Moreover, irAEs occurred earlier in overweight/obese patients than in normal weight/underweight patients, although without any difference in severity.

The mechanisms of such an intriguing association between BMI, a surrogate measure of body fat, and ICI therapies are not completely understood. Obesity is a low-grade inflammatory metabolic condition that has been associated with both cancer and autoimmunity (42, 43). Indeed, obesity is associated with increased adipose tissue, metabolic disturbances (hyperglycemia), higher levels of insulin, and insulin-like growth factors (IGFs), with potent mitogenic activity (44–46). Moreover, it has been associated with increased secretion by adipocytes of pro-inflammatory cytokines (TNF-α, IL-6 and IL-1β) and adipokines (leptin, adiponectin, resistin), which could affect T cell function, resulting in Th1/Th2 imbalance and promoting a pro-inflammatory state (47, 48). Such a pro-inflammatory condition is well known to predispose to the occurrence of autoimmune disorders (49, 50); hence, excess body weight may promote the development of irAEs. Moreover, fat accumulation leads to enhanced infiltration of pro-inflammatory CD8⁺ T cells into adipose tissue, accompanied by a reduction in adipose-resident regulatory T cells (Tregs) (51). Additionally, other obesity related factors such as dietary habits, genetic susceptibility, and microbiome may contribute to increased occurrence of irAEs in obese patients.

Worthy of note, our study is the first to suggest that irAEs occur more frequently in overweight/obese patients with a concurrent dysmetabolic state, defined as the presence of hypertension and/or glycemic disturbances and/or dyslipidemia, suggesting that low grade meta-inflammation, well known to be associated with obesity and related metabolic disorders, may represent a predisposing condition for development of irAEs in patients with higher BMI. Therefore, baseline BMI and related dysmetabolic conditions should be considered among the potential risk factors for the development of irAEs, along with other potential predictors, such as a family or personal history of autoimmune disorders, the use of immunotherapeutic combinations or previous TKI treatment. This would help clinicians in identifying patients who are at higher risk for irAEs, thereby personalizing therapeutic choices and clinical monitoring.

Alternatively, overexposure to treatment may occur in overweight patients due to an increased dose calculation based on mass weight (52). In this light, sarcopenic obesity, a not rare condition in oncologic patients, represents a complex and emerging factor in cancer patients undergoing ICI therapy. Characterized by the coexistence of low muscle mass and excess adiposity, it has been associated with increased toxicity and poorer clinical outcomes in various cancer settings (53–55). Recently, a systematic review and meta-analysis examining the impact of sarcopenia on cancer patients treated with ICIs found that sarcopenia is associated with an increased risk of irAEs, though the relationship with irAEs was less clear (56). Moreover, sarcopenic obesity may act as a confounding factor, influencing both the incidence of irAEs and treatment efficacy, complicating the interpretation of clinical outcomes (53). This dual role underscores the importance of considering body composition, beyond simple measures of body weight, when interpreting clinical associations. Further studies are needed to elucidate its precise impact and underlying mechanisms.

Even more complex is the relationship between obesity, cancer outcomes and response to cancer treatment in the context of ICI treatment. Obesity has been recognized as a risk factor and a negative prognostic factor for several cancers, worsening oncological outcomes, including recurrences, disease-free survival, all-cause and cancer specific mortality (29). Nevertheless, evidence suggests that overweight and obesity may be associated with better oncological outcomes than normal weight, and an inverse relationship between BMI and mortality (the so-called ‘obesity paradox’) has been found in several cancers at advanced stage, although the underlying mechanisms are not fully understood (29, 50).

A possible suggested mechanism could be the better objective responses to immunotherapy observed in obese compared to non-obese patients, with significantly longer PFS and OS (57). A high BMI has been associated with better clinical outcomes to ICI therapy. Complex interactions between the adipose tissue and tumor cells have been described (58). The modulation of the tumor microenvironment by obesity-associated molecules, including hormones and pro-inflammatory cytokines (e.g. TNF-α, IL-6) can play a key role in promoting tumor development and progression and, at the same time, in enhancing T-cells function and immune responses to ICIs. Besides inflammation, other mechanisms can modulate the effects of obesity in cancer patients, including alterations of insulin-like growth factor pathways, induction of hypoxia and HIF-1α signaling, and modulation of microbiota (59). The role of BMI as a predictor of toxicity from anti-neoplastic drugs should be further explored. In advanced cancer patients treated with ICIs, several studies reported better outcomes, in terms of longer PFS and/or OS, in overweight/obese patients compared to patients with normal BMI, with differences by sex (26, 27, 60, 61).

In our cohort, we failed to find any relationship between obesity, occurrence of irAEs and treatment efficacy in terms of either PFS and/or OS. However, it is widely known that survival outcomes can be influenced by several factors related to patients and/or cancer. We hypothesize that this may be due to several factors. First, the sample size of the study was limited, and our patient cohort included various tumor types and stages — although the majority were advanced — which can independently be associated with different prognoses, regardless of treatment. Moreover, since this was a real-world, observational study, patients with some significant comorbidities (i.e., cardiovascular) and with ECOG PS 2, who are usually excluded from large registration trials, were also included. Finally, other important factors could have potentially affected survival outcomes, including previous therapeutic lines and the great variability among patients in the timing and methods used for disease response evaluation, as for clinical practice. However, we decided to assess PFS and OS because these outcomes can better reflect the long-term efficacy of immunotherapy compared to treatment response.

Some limitations should also be noted, including: (i.) the retrospective design; (ii.) the relatively limited number of patients; (iii.) the inclusion of patients with different types of cancer, introducing some clinical heterogeneity and potential biases. Interestingly, our cohort of 130 patients had no grade 3/4 irAEs or irAEs leading to treatment discontinuation, which is lower than would be expected according to literature data (25, 27). In the multi-center retrospective study by Cortellini and coworkers, including 1070 advanced cancer patients treated with PD-1/PD-L1 inhibitors, higher BMI was significantly related to higher occurrence of G3/G4 irAEs and therapy discontinuation (25). The absence of high-grade irAEs in our series may be due to the small sample size compared to larger studies, and/or to recruitment bias. Indeed, our study was first designed to assess endocrinological AEs, hence we excluded patients with a previously diagnosed thyroid dysfunction that could potentially represents an important predisposing factor for other irAEs. Endocrinological AEs, the most common type of irAEs, are usually low grade, and, if promptly diagnosed and treated in the context of an experienced team of endocrinologists and oncologists, do not worse and/or lead to treatment discontinuation. Overall, this should be acknowledged as a limitation of our work, preventing us from conducting a more in-depth investigation into the relationship between BMI and the severity of irAEs, as demonstrated in other studies.

Overall, major strengths of the study are: (i.) access to complete information (hospital-based data) regarding patients at baseline and during ICI-treatment; (ii.) a real-life scenario, that assesses irAEs presentation and management in regular clinical practice, thereby reflecting real adherence to treatment/intervention and outcomes; (iii.) a homogeneous cohort of patients belonging to the same geographical area followed-up at a single center. Thus, this real-life study provides evidence on how treatments perform in routine clinical practice, capturing a broader and more heterogeneous patient population than controlled trials. It offers valuable insights into effectiveness, safety, and feasibility in everyday care, complementing existing literature. We have to acknowledge that the small number of patients limits subgroup analyses and precludes drawing definitive conclusions, while noting that the observed trend of higher irAE rates in overweight/obese patients with metabolic comorbidities remains an interesting finding. Further studies on larger series are needed to verify whether these results can be extrapolated to other populations and confirmed on large series.

In conclusion, despite advances in the knowledge of the peculiar profile of toxicity of ICI-based therapies, many questions regarding irAEs remain to be fully addressed. These include the role of predisposing factors like BMI and gender, as well as the possible association between occurrence of irAEs and response to ICI treatment. Also, clinical and biochemical predictors of the risk for developing irAEs are needed. In our well-characterized cohort of patients treated with ICIs, we confirmed that overweight/obesity is associated with increased risk of irAEs, with a notable predictive value, mostly when accompanied by dysmetabolic conditions. However, no clear association between BMI and immunotherapy efficacy was observed, in terms of either PFS or OS. These results may help oncologists to identify the patients who are most likely to develop irAEs, improving the management of their patients in a real-life scenario.