Alpelisib-induced hyperglycemia represents the most prevalent adverse event associated with treatment with this drug, occasionally posing a challenge to treatment continuity. Our study demonstrates that a collaborative intervention between a Diabetes Unit and an Oncology Department enables effective glycemic control during follow-up, thereby averting the need for treatment discontinuation due to this cause. Furthermore, our findings identify C-peptide and BMI as factors associated with an increased need for anti-hyperglycemic pharmacological interventions.

Some analysis of pivotal studies underscored the significance of pre-treatment glucose, HbA1c levels [11], BMI age and monocytes in the incidence of hyperglycemia as an adverse event of Alpelisib treatment [12]. Consequently, the need for increased glycemic monitoring to mitigate the clinical impact of hyperglycemia and the discontinuation of oncological treatment has been recognized [6]. Therefore, implementing protocols and seeking guidance from a diabetologist has been recommended for patients scheduled for Alpelisib treatment [10, 13].

Along these lines, the joint follow-up between the Diabetes Unit and the Oncology Department of our center resulted in a lower incidence of moderate or severe hyperglycemia (grades 3 and 4) and consequently a reduced incidence of dose reduction, temporary suspension, or definitive discontinuation of Alpelisib compared to other studies [5, 11, 13]. These favorable outcomes might be attributed to early training sessions on management of hyperglycemia at treatment onset and closer glycemic monitoring in these patients, as previously described in other reports [13].

Several sociodemographic, clinical, or laboratory parameters such as age, BMI, fasting plasma glucose, and HbA1c have been identified as factors capable of predicting the onset and severity of hyperglycemia in patients treated with PI3K inhibitors [12, 14]. In our series, fasting glucose and HbA1c before treatment with Alpelisib did not exhibit sufficient predictive power. This discrepancy might be explained by the absence of carbohydrate metabolism alterations in the majority of our patients and the different characteristics of the study populations.

Insulin and C-peptide are co-secreted by the pancreas in an equimolar ratio. Unlike insulin, C-peptide undergoes minimal hepatic degradation, and its plasma concentration follows a linear kinetics. Therefore, plasma C-peptide levels have been suggested to be more suitable than plasma insulin levels to monitor pancreatic insulin secretion. Consequently, C-peptide determination is valuable to differentiate between T1D and T2D [15]; moreover, C-peptide levels are associated with the need for insulin therapy in T2D [16]. Elevated C-peptide levels are also linked to insulin resistance and BMI [17]; furthermore, there is speculation regarding the potential independent role of C-peptide in atherosclerosis initiation [18]. However, C-peptide involvement in Alpelisib-induced hyperglycemia had not been explored before the current study. Our study establishes a risk profile based on C-peptide and BMI baseline levels. Defining a risk phenotype may allow planning more intensive diabetes monitoring strategies aimed at high-risk patients, such as continuous glucose monitoring. This system has been useful to understand Alpelisib pharmacodynamics and has been a valuable tool for managing and controlling hyperglycemia [19]. Predictive tools like the risk profile we propose could aid in identifying patients more likely to develop significant hyperglycemia, thereby facilitating the selective prescription of glucose-lowering medications in cases where it is truly necessary.

This study has several limitations. Firstly, its retrospective design limits the ability to establish causality, only allowing to merely propose hypotheses. The absence of a control group complicates the thorough verification of the intervention impact on patient follow-up. Additionally, the majority of patients in the study discontinued treatment due to disease progression, thus hampering medium-term Alpelisib follow-up. In addition, our study is limited to a small population from a single hospital. These limits cannot guarantee the generalizability of the results obtained in the present study, Thus, no general messages can be drawn but only preliminary data or suggestions. Therefore, further studies with larger and more diverse cohorts should be conducted before the risk factors identified in our study could be extrapolated to other populations, especially to populations of Asian descent, where the impact of BMI and insulin resistance differs from that of populations of Caucasian descent [7, 20].

In summary, hyperglycemia is a highly prevalent adverse event in women with metastatic breast cancer treated with Alpelisib. Follow-up coordinated by both Diabetes and Oncology Units can effectively manage glycemic levels and therefore prevent drug discontinuation due to this cause. Baseline C-peptide and BMI at the onset of treatment can be associated with an increased need for anti-hyperglycemic treatment. Given the methodological limitations of this study, future prospective multicenter studies should be conducted to verify the hypotheses proposed here.