The phosphoinositide 3-kinase (PI3K) pathway is frequently altered in human cancer, reported at a rate of nearly 30% in solid cancer (1–3). Aberrant activation of this pathway is associated with cancer cell proliferation, cell survival, and angiogenesis (4–6). The PIK3CA gene (the phosphatidylinositol-4,5bisphosphate 3-kinase catalytic subunit alpha gene) is the most frequently altered PI3K isoform in solid tumors, with gain-of-function mutations upregulating downstream AKT-mTOR signaling pathways that promote cancer cell growth and proliferation (7, 8). Alpelisib, an oral selective inhibitor of class I PI3K p110α (9, 10), has demonstrated antitumor properties in preclinical studies, particularly in the PIK3CA-mutated cancer models (10). In a first-in-human phase I study with advanced solid malignancies with PIK3CA alterations, alpelisib showed encouraging preliminary activity and a tolerable safety profile, with the MTD declared as 400 mg once daily (11). However, the clinical efficacy of single-agent alpelisib has been limited, suggesting that combination therapy may be more effective. Alpelisib has been combined with multiple anticancer drugs, such as fulvestrant (12, 13), everolimus (14), olaparib (15), encorafenib (16), paclitaxel (17), cisplatin (18), and cetuximab (19). Based on the results of the SOLAR-1 trial, alpelisib has been approved in combination with the fulvestrant for the treatment of PIK3CA-mutated, HR(hormone receptor)-positive, HER2(human epidermal growth factor receptor 2)-negative advanced breast cancer who had received endocrine therapy previously (12). Currently, Alpelisib is being studied in different combinations, expecting to improve synergy effects and overcome resistance to chemotherapy.

Capecitabine is an oral prodrug of 5-FU (fluorouracil) that is effective as chemotherapy treatment for multiple cancers, including colorectal (20, 21), gastric (22), pancreatic (23), and breast cancer (24). As a monotherapy, it has been widely used for salvage regimens in Korea based on retrospective data with a moderate disease control rate of 60% (25). Capecitabine has demonstrated safety and anti-tumor activity when used in combination with other chemotherapy or biological agents such as bevacizumab and lapatinib (26, 27). Additionally, the combination of capecitabine with an mTOR inhibitor targeting the PI3K/Akt/mTOR signaling pathway has shown antitumor effects and tolerable safety (28, 29). Preclinical data has also suggested that combining PI3K inhibition with 5-fluorouracil could increase the anticancer effects in cancer cell lines (30).

In this phase I trial, we evaluated the combination of capecitabine and alpelisib in patients with advanced solid cancer to determine its safety, tolerability, and pharmacokinetic interaction. This phase I trial aimed to assess whether this combination therapy would be a feasible and effective treatment option, especially for patients with PIK3CA mutant tumors.

This phase Ib study explored the safety, MTD, and PR2D of the combination of capecitabine with the oral PIK3CA inhibitor alpelisib in patients with advanced solid tumors refractory to standard therapy. The results showed that the combination treatment was generally well-tolerated and had an acceptable toxicity profile. Only one DLT, which was grade 3 hyperglycemia, was observed in one out of six patients treated at dose level 1. Based on our study’s definition, we identified the MTD and RP2D as alpelisib 300 mg daily continuously combined with 1,250 mg/m² capecitabine twice daily for 14 days in 3-weekly cycles. Therefore, we planned further phase II trials to evaluate the efficacy and safety of this combination therapy using the established dosing regimen.

Hyperglycemia was the most commonly reported treatment-related adverse event and the only dose-limiting toxicity observed in our study. Hyperglycemia is a well-known side effect of PI3K-AKT-mTOR pathway inhibitors. The inhibition of this pathway can lead to abrogated insulin function, impaired insulin secretion, and the development of insulin resistance, resulting in hyperglycemia (2, 4, 31). Previous studies have reported the incidence of hyperglycemia associated with PI3K-AKT-mTOR pathway inhibitors in large phase III clinical trials ranging from 13 to 63.7% and high-grade hyperglycemia from 4 to 32.7% (32–34). The incidence of hyperglycemia may also differ depending on the combination drug, disease stage, and alpelisib dose used. For example, in the phase 1b study of triple-combination therapies (encorafenib and cetuximab with alpelisib) in colorectal cancer (BRAF-mutant), the incidence of hyperglycemia was 39.3% (16), while a phase II study of alpelisib plus fulvestrant in hormone receptor-positive breast cancer resulted in 29% of hyperglycemia (35). In a phase III study (SOLAR-1), hormone receptor-positive breast cancer showed a higher incidence of hyperglycemia at 63.7% (12). While the frequency of hyperglycemia by cancer type is not well-known in previous clinical studies, it is clear that the severity and incidence of hyperglycemia can vary based on several factors.

In our study, all-grade hyperglycemia was reported in 66.7% (2/3) of patients in the alpelisib 100 mg BID group (dose level 0) and 77.8% (7/9) of patients in the alpelisib 150 mg BID group (dose level 1,2). Grade 3 hyperglycemia was observed in 16.7% (2/12) of patients. The median time to onset of all-grade hyperglycemia was 3.1 weeks (range, 1.0 to 8.7). Although the number of patients was small, there appeared to be a difference in hyperglycemia incidence according to cancer type. Specifically, hyperglycemia was reported in all colon cancer patients and in half of the breast cancer patients. In our study, colon cancer patients appeared to be at a higher risk of hyperglycemia, likely due to the presence of several risk factors such as old age, history of bowel resection, and use of medication that can increase the risk of hyperglycemia. The median age of colon cancer patients was higher than that of breast cancer patients (58.5 years vs. 49.5 years), and all colon cancer patients underwent partial colorectal resection. Furthermore, most colorectal cancer patients were enrolled as third-line chemotherapy. They received FOLFOX (folinic acid, fluorouracil, and oxaliplatin) and FOLFIRI (folinic acid, fluorouracil, and irinotecan)-based chemotherapy in the previous first and second-line treatments, which may have made them more vulnerable to hyperglycemia due to long exposure to dexamethasone with standard pre-medication. Therefore, our findings suggest that clinicians should exercise caution when administering alpelisib and capecitabine to colorectal cancer patients with risk factors for hyperglycemia and closely monitor blood glucose levels during treatment. Further studies are needed to evaluate the safety of this combination therapy using the established dosing regimen.

Other adverse events of combination therapy were nausea, stomatitis, and hand-foot syndrome. The hand-foot syndrome can be attributed to capecitabine since this has not been observed before in single-agent alpelisib trials. This well-known side effect of capecitabine resulted in dose reductions of capecitabine in our patients. As previously known in the study, the mean time until all grades of hand-foot syndrome occurred was 6.1 weeks (range, 3.1 to 15.0), as was previously known from the study. Although stomatitis is a common adverse event of both capecitabine and alpelisib as a single agent, this overlapping toxicity remained mild to moderate in severity in this study and was not dose-limiting.

The combination of alpelisib and capecitabine demonstrated sustained clinical activity in some breast cancer patients with PIK3CA mutations. Two patients with breast cancer are still on the study for over 12 months Median progression-free survival (PFS) of 10.4 months showed promising results in patients with heavily-treated breast cancer. In particular, three patients had more than 10 cycles of treatment and had the following mutations: PIK3CA H1047L, PTEN 317_318del, PTCH1 D898N, CDH1 S649fs; PIK3CA E545K, PIK3CA E726K, CDH1 R598*, AR P549S; and PIK3CA N345K TP53 M237I, respectively, which PIK3CA mutation points were heterogeneous. Although this studies had small sample size, previous studies have shown that CDH1 mutation might enhance the effectiveness of PIK3CA inhibitors when present together (36, 37).

This study did not show significant clinical benefits in colorectal cancer patients compared to breast cancer. This suggests that the role of PIK3CA mutation as a driver mutation may vary depending on the cancer type. The median PFS for this treatment was only 1.4 months. This could be due to the absence of PIK3CA mutation in half of colorectal cancer patients, and the frequent co-occurrence of KRAS pathogenic variants with PIK3CA mutations, leading to a shorter reported PFS and potentially reduced activity against PIK3CA inhibitors. Studies have shown that tumors with aberrant activation of the RAS/RAF/MEK/ERK pathway, such as KRAS-mutant cancers, do not respond to PI3K pathway inhibitors (38–40). One study of pictilisib (GDC-0941) found that patients with both PIK3CA and KRAS mutations had a lower response rate and shorter progression-free survival compared to those with only a PIK3CA mutation (41). However, further research is needed to confirm and expand upon these findings.

The results of our study demonstrate that the pharmacokinetics of alpelisib and capecitabine are not affected when these agents are co-administered in patients with advanced cancer. In particular, the Cmax and AUC values of both drugs were similar across all dose levels, indicating that there were no pharmacokinetic interactions between alpelisib and capecitabine. Our study also found that the time to reach peak plasma concentration (tmax) was consistent with previous reports of alpelisib and capecitabine. These findings suggest that it is safe to combine these two agents in patients with advanced cancer. Overall, the pharmacokinetic results of our study support the safety and tolerability of the alpelisib-capecitabine combination and provide a basis for further investigation of this combination in larger phase II clinical trials. Additional pharmacokinetic analyses may be useful in these trials to further characterize the pharmacokinetics of alpelisib and capecitabine and to optimize dosing strategies for this combination therapy.

Limitations of this study include the small number of patients, no control groups, and data for a single institution. In addition, since our study only enrolled breast and colorectal cancer patients, our findings may not be generalizable to all solid tumors, and further studies are needed for other types of cancer. Although our study has reached several endpoints, definitive conclusions cannot be drawn with respect to the synergistic effects of this combination. Moreover, we currently lack clear biomarkers to predict which patients have long-term responses. In addition, insufficient pharmacokinetics and pharmacodynamic analyses are requested better to understand the effects of dose reduction and discontinuation.

In conclusion, our study demonstrated that the combination of alpelisib 300 mg daily and capecitabine 1,250 mg/m² twice daily for 14 days in 3-weekly cycles is a safe and well-tolerated treatment option for patients with advanced cancers harboring PIK3CA mutations. Our findings also indicate that this combination therapy achieved prolonged clinical benefits in a significant number of patients. Notably, toxicities associated with this treatment were generally manageable, and no unexplained severe toxicities were reported.

The current therapies for PIK3CA mutant advanced solid tumors are being explored in variable drug combinations. A larger prospective study and following experiment should be conducted to elucidate the role of PIK3CA mutations in solid tumors. Overall, our study suggests that the combination of alpelisib and capecitabine could be a promising treatment option for patients with PIK3CA mutant advanced cancers. Larger studies to validate our results and further investigation is needed in phase II trial.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

The studies involving humans were approved by the institutional review boards of Korea University Anam Hospital (2020AN0539) and by the FDA. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

AL: Data curation, Formal analysis, Writing – original draft, Writing – review & editing, Conceptualization, Methodology, Visualization. BK: Formal analysis, Writing – review & editing. JK: Data curation, Writing – review & editing. MH: Data curation, Writing – review & editing. KP: Conceptualization, Writing – review & editing. YK: Conceptualization, Writing – review & editing. SL: Conceptualization, Data curation, Supervision, Writing – original draft, Writing – review & editing.