First Experiences with Alpelisib in Clinical Routine: Case Reports from a German Breast Center

Abstract

Introduction

The phosphatidylinositol-3-kinase (PI3K) inhibitor alpelisib is the only approved agent for treating ­PIK3CA-mutated, hormone receptor-positive (HR+) human epidermal growth factor receptor 2-negative (HER2−) advanced breast cancer (ABC). Trials have reported hyperglycemia, diarrhea, and rash as the main grade 3 side effects.

Methods

In a managed access program (ClinicalTrials.gov ID: NCT03706573; start: 06/2019), 8 HR+ HER2– ABC patients with a median 4.5 prior therapy lines were treated with alpelisib at the Breast Center of the Ludwig-Maximilian University (LMU) Hospital, Munich, based on the results of a new-generation sequencing (NGS) panel and PIK3CA mutation analysis by the Molecular Tumor Board of the Comprehensive Cancer Center, Munich.

Results

Median therapy duration was 3.42 months for patients who discontinued and 3.95 months for those still on alpelisib (4 pts). Five had hyperglycemia (1 with grade 3) with fasting glucose levels of up to 450 mg/dL that required hospitalization and insulin therapy. Two experienced rash (grades 1 and 3) and 2 reported grade 3 diarrhea. Supportive therapy as well as interruption and/or dose reduction were necessary to control treatment-associated side effects.

Conclusion

Patient education and a well-trained, interdisciplinary team including diabetologists, from the initiation of alpelisib treatment onwards, are essential to safely treat ABC patients with this new drug and to maintain their quality of life and ensure their survival.

Introduction

The treatment of hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2−) advanced breast cancer (ABC) has undergone substantial changes in recent years. The addition of cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) to endocrine therapy (ET) has led to improved progression-free survival (PFS) and longer overall survival (OS) [1, 2, 3, 4, 5]. This has become the standard of treatment for HR+ HER2– ABC [6]. However, resistance to ET is still a challenge [7].

A known mechanism of endocrine resistance is hyperactivation of the α-isoform of phosphatidylinositol-3-kinase (PI3K) caused by activating mutations in the PI3K catalytic subunit α gene (PIK3CA) [8]. Activating mutations occur in approximately 40% of all HR+ HER2– ABC cases [9, 10]; α-selective PI3K inhibitors have been developed and recently investigated in clinical trials. Alpelisib is an orally available PI3K inhibitor, which has shown effectiveness in PIK3CA-mutated tumors, especially in combination with fulvestrant [11, 12]. Two recent major studies have investigated the efficacy and safety of adding alpelisib to ET in patients with pretreated HR+ HER2– ABC. In the phase 3 SOLAR-1 trial, fulvestrant plus alpelisib/placebo was administered to 571 patients (572 overall, but 1 was never treated) with HR+ HER2– ABC (in 2 separate cohorts, i.e., with or without PIK3CA-mutated tumors), who had previously undergone ET. In patients with tumor-tissue PIK3CA mutations (n = 341), the addition of alpelisib led to a significantly longer PFS (11.0 months in the alpelisib + fulvestrant vs. 5.7 months in the placebo + fulvestrant group, p < 0.001), improved overall response (OR) rates (26.6 vs. 12.8%) and a numerically, but not statistically significant increase (7.9 months) in OS (39.3 vs. 31.4 months) [13, 14]. Alpelisib did not significantly influence PFS and OR in the patients without PIK3CA mutations [13]. However, only 5.3% of the patients in PIK3CA-mutated cohort of the SOLAR-1 trial who received alpelisib were pretreated with CDK4/6i, which is the current standard of care. The ongoing phase 2 BYLieve trial is evaluating alpelisib and ET in patients with HR+ HER2– PIK3CA-mutated ABC who progressed on prior therapy. Patients were categorized according to the immediate prior therapy into 3 cohorts: prior CDK4/6i + aromatase inhibitor (AI), prior CDK4/6i + fulvestrant, and prior ET or chemotherapy. The first results in the prior CDK4/6i + AI cohort were demonstrated at ASCO 2020 and showed a median PFS of 7.3 months, with 50.4% of the patients alive without progression after 6 months. Alpelisib thus seems to have meaningful efficacy, even after prior CDK4/6i therapy [15, 16]. Both trials report manageable toxicity of alpelisib [13, 16].

The consistent and significant results of these 2 trials led to the FDA approval of alpelisib for the treatment of HR+ HER2– PIK3CA mutated ABC in May 2019 [17]. The European Medicines Agency (EMA) granted marking authorization on 27 July 2020 [18]. Real-world data about the toxicity and therapy management of alpelisib are limited due to its relatively recent approval. At the Breast Center of the Ludwig-Maximilian University (LMU) Hospital, Munich, which is a part of the Comprehensive Cancer Center (CCC) Munich, we had the opportunity to treat 8 patients with HR+ HER2– ABC with alpelisib as part of a managed access program (MAP) before alpelisib was approved by the EMA, and it became commercially available in Germany on 1 September 2020. Here, we want to present our real-world experience with this drug, compare side effects during the MAP to the toxicity described in the clinical studies, and give advice for the management of patients based on our clinical experience.

Materials and Methods

Requirements for Clinical Implementation

Essential for the early implementation of alpelisib in clinical treatment were the structures provided at the CCC Munich, which has provided a so-called Molecular Tumor Board for patients with heavily pretreated cancer since 2018. As part of this board, next-generation sequencing (NGS) and analysis of known key mutations are performed in the tumor tissue of these patients. An interdisciplinary panel discusses the molecular results, analyzes, and recommends (if possible) mutation-based rather than tumor entity-based therapies. The PIK3CA mutation analysis was well-established and an extensive study network was available. The MAP to provide alpelisib for patients with PIK3CA-mutated ABC (ClinicalTrials.gov ID: NCT03706573) was available at the CCC Munich from October 2019 and ended with the commercial availability of alpelisib in Germany in September 2020. The MAP was available for patients with HR+ HER2– ABC and proven tumor-tissue PIK3CAmutations. Only postmenopausal patients who had exhausted all standard ET regimes could be included. To be eligible for inclusion in the MAP, patients had to have experienced progression under all lines of ET therapy with AI and tamoxifen as well as under both CDK4/6i and everolimus, and they were therefore heavily pretreated. Patients with preexisting diabetes mellitus type 1 (DM1) or not-controlled DM2 based on fasting plasma glucose and HbA_1c levels were excluded. Detailed inclusion and exclusion criteria are displayed in online supplementary Table 1 (www.karger.com/doi/10.1159/000514794).

Comparison of SOLAR-1 Trial, BYLieve Trial, and Real-World Adverse Events

Both the SOLAR-1 trial and the preliminary analysis of the BYLieve trial describe manageable toxicity of alpelisib. Adverse events (AEs) that occurred in at least 15% of all patients of the alpelisib + fulvestrant cohorts in 1 of these trials appear in Table 1. Common AEs of any grade (SOLAR-1/BYLieve) were hyperglycemia (in 63.7/58% of all patients), diarrhea (57.7/60%), nausea (44.7/46%), decreased appetite (35.6/28%), and rash (35.6/28%). The most common grade 3 AEs were hyperglycemia (in 32.7/28.3% of all patients), rash (9.9/9.4%), and diarrhea (6.7/5.5%) [13, 16].

Table 1.

Adverse events in the alpelisib + fulvestrant cohorts of the SOLAR-1 and BYLieve trials and in patients included in the MAP at the Breast Center of the LMU Munich (adapted from [13, 15])

AE	SOLAR-1 trial: alpelisib + fulvestrant group (n = 284)		BYLieve trial: alpelisib + fulvestrant in the prior CDKi + AI cohort (n = 127)		Breast Center LMU Munich: alpelisib + fulvestrant in MAP (n = 8)
	all	≥ grade 3	all	≥ grade 3	all	≥ grade 3
Hyperglycemia	181 (63.7)	93 (32.7)	74 (58.3)	36 (28.3)	5 (63.5)	1 (12.5)
Diarrhea	164 (57.7)	19 (6.7)	76 (59.8)	7 (5.5)	2 (25)	2 (25)
Nausea	127 (44.7)	7 (2.5)	58 (45.7)	0	1 (12.5)	1 (12.5)
Fatigue	69 (24.3)	10 (3.5)	37 (29.1)	1 (0.8)	2 (25)	0
Decreased appetite	101 (35.6)	2 (0.7)	36 (28.3)	1 (0.8)
Rash	101 (35.6)	28 (9.9)	36 (28.3)	12 (9.4)	2 (25)	1 (12.5)
Stomatitis	70 (24.6)	7 (2.5)	34 (26.8)	2 (1.6)
Vomiting	77 (27.1)	2 (0.7)	30 (23.6)	2 (1.6)	1 (12.5)	1 (12.5)
Asthenia	58 (20.4)	5 (1.8)	25 (19.7)	1 (0.8)
Headache	50 (17.6)	2 (0.7)	24 (18.9)	1 (0.8)
Pruritus	51 (18.0)	2 (0.7)	20 (15.7)	2 (1.6)	1 (12.5)	0
Dry skin			20 (15.7)	1 (0.8)	1 (12.5)	0
Dyspnea			19 (15.0)	3 (2.4)
Weight loss	76 (26.8)	11 (3.9)			1 (12.5)	0
Alopecia	56 (19.7)	0
Mucosal inflammation	52 (18.3)	6 (2.1)
Dysgeusia	47 (16.5)	0

Values express n (%). AEs that were reported in the trials in at least 15% of all patients in the subgroups are displayed. As not all AEs were reported in both trials, some fields remain empty. The most common grade 3 AEs are displayed in italics. AEs at the Breast Center of the LMU Munich were not obtained under trial conditions but in daily clinical practice. AEs, adverse events; MAP, managed access program.

Results

Implementation of Alpelisib in Clinical Treatment at the CCC Munich and Treatment Duration

Eight patients were treated with alpelisib at the Breast Center of the LMU Munich within the MAP. Four patients stopped alpelisib therapy with a treatment duration of 98–132 days (median 104 days or 3.42 months). The reason for stopping alpelisib was disease progression in all 4 cases. Four patients were still under therapy at the end of the MAP. Treatment duration in patients with ongoing therapy ranged from 7 to 249 days in the MAP (median 120 days or 3.95 months, as measured on 31 August 2020; Table 2). Among these patients, 2 showed partial regression and 1 had stable disease. One patient started alpelisib only shortly before the cut-off date for data collection, so information about efficacy is not yet available. Median patient age was 64.5 years (range 42–72 years; Table 2). Two younger patients (42 and 47 years) had a history of bilateral salpingo-oophorectomy. Our patient cohort was heavily pretreated with a median of 4.5 (range 2–11) prior therapy lines for ABC (Table 2).

Table 2.

Patient characteristics and duration of alpelisib therapy at the Breast Center of the LMU Munich

Patient No.	Age at baseline, years	Number of prior therapy lines for ABC	Therapy ongoing	Therapy duration, days
				therapy stopped	ongoing therapy
1	72	10	No	132
2	47a	8	No	109
3	70	11	Yes		249
4	42a	3	No	98
5	72	5	Yes		173
6	59	3	No	99
7	58	4	Yes		67
8	59	2	Yes		7
Median	64.5	4.5		104	120

For patients with ongoing therapy, the MAP ended in September 2020 and treatment duration in the MAP was measured until 31 August 2020.

^aPremenopausal women with a history of bilateral salpingo-oophorectomy.

AEs under Alpelisib Therapy at the LMU Breast Center

The observed AEs in our patient cohort compared to the reported AEs are displayed in Table 1 (right). The AEs chiefly observed in our cohort were hyperglycemia, diarrhea, and rash.

Hyperglycemia (details in Table 3) occurred in 5 of our 8 patients (62.5%). One (12.5%) experienced grade 3 hyperglycemia (defined as fasting plasma glucose or random blood glucose >250–500 mg/dL [>13.9–27.8 mmol/L] according to CTCAE v4.03 [19]. In our hyperglycemic patients, the maximum fasting blood sugar level after starting alpelisib ranged from 117 to 450 mg/dL and the maximum HbA_1c from 5.4 to 7.8%. The single case of grade 3 hyperglycemia occurred already after 7 days of therapy with alpelisib, with a maximum fasting blood sugar level of 450 mg/dL. The patient was hospitalized and treated with insulin therapy; alpelisib treatment was interrupted for 15 days and the dose was then reduced. Grade 3 hyperglycemia did not recur. In the 4 patients with grade 1–2 hyperglycemia, blood sugar levels rose slowly over time, with fasting blood sugar levels elevated for the first time after 15–103 days of therapy. In these patients, no treatment interruptions or dose reductions were necessary. All patients were referred to a diabetologist. Initially, we transferred patients to a diabetologist only when blood sugar levels were rising; later, we included a diabetologist in the interdisciplinary treatment team from the beginning of the alpelisib therapy. We prescribed blood glucose monitors to all patients and instructed them to measure blood sugar levels regularly. We further informed patients about the symptoms of hypo- and hyperglycemia and ketoacidosis.

Table 3.

Hyperglycemia due to alpelisib therapy at the LMU Munich

Patient No.	Hyperglycemia
	occurrence	day of occurrence	maximum fasting blood sugar, mg/dL	maximum HbA1c, %
1	Yes	103	163	7.3
2	Yes	7	450	7.8
3	No
4	Yes	54	117	5.4
5	Yes	85	135	6.1
6	Yes	15	122	6.2
7	No
8	No

Occurrence of hyperglycemia, the first day at which elevated blood sugar levels were measured, and the maximum values for fasting blood sugar and HbA_1c are displayed.

Rash occurred in 2 patients (25%). One (12.5%) experienced generalized rash (CTCAE grade 3 [19]), which occurred after 8 days of alpelisib treatment and was managed with therapy interruption, antihistamines, and cortisone therapy. After a dose interruption of 20 days, alpelisib was resumed at a reduced dose. The other patient experienced a grade 1 rash on day 15 of alpelisib treatment and was managed with antihistamines and local cortisone. Both patients were referred to a dermatologist. Another patient did not experience rash but had mild pruritus during her first days on alpelisib. All patients were informed about rash as a possible side effect. They received a prescription for antihistamines at the initiation of alpelisib, were told to take nonsedating antihistamines (cetirizine 10 mg) if rash or itching occurred and to suspend the alpelisib therapy at this time.

Diarrhea occurred in 2 of our 8 patients (25%) after 11 and 13 days of alpelisib therapy, respectively. Both were cases of grade 3 diarrhea (according to CTCAE: increase of ≥7 stools/day over baseline, hospitalization indicated, severe increase in ostomy output compared to baseline, limiting self-care ADL) [19]. One of these patients also suffered from severe nausea, vomiting, dehydration, and weight loss; she was hospitalized and treated with intravenous fluids, loperamide, and alpelisib interruption for 16 days followed by a dose reduction. In the other patient, outpatient treatment with increased fluid intake, loperamide, and alpelisib interruption for 8 days, followed by dose reduction was possible. At initiation of alpelisib therapy, all patients were informed about the possibility of diarrhea and received loperamide as an emergency medication.

Other side effects were reported by our patients (Table 1). One with prior capecitabine use experienced very dry skin that was treated with hemp oil ointment and urea-containing lotions. A second patient was hospitalized due to abdominal pain of unknown etiology that was treated symptomatically. This pain was most likely related to prior radioembolization therapy of the liver. Two patients reported fatigue. Another had arterial hypertension that was managed medically from day 77 of alpelisib therapy onwards.

Discussion

Recommendations for Therapy Management Based on Real-World Experience with Alpelisib

To evaluate a patient's eligibility for alpelisib treatment, the PIK3CA mutation status must be determined in a tumor-tissue sample, preferably obtained during the last recent progression or as a liquid biopsy. We were able to offer PIK3CA testing as part of our Molecular Tumor Board at the CCC Munich, which was essential for our participation in the alpelisib MAP. In our patient cohort treated within the MAP, treatment duration (median 3.42 months in stopped and 3.95 months in ongoing therapies at the end of the MAP) was shorter than in the SOLAR-1 trial (median 5.5 months [13]). However, due to the MAP inclusion criteria, our patient cohort was more heavily pretreated (median 4.5 prior therapy lines for ABC) than the SOLAR-1 trial patients (alpelisib + fulvestrant cohort in PIK3CA-mutated breast cancer: 52.1% first line and 46.7% second line [13]).

In our real-world setting, we saw side effects of alpelisib treatment comparable to those described in the SOLAR-1 and BYLieve trials, even though patients had been heavily pretreated. The observed side effects were manageable with early interventions, sufficient supportive measures, therapy interruption, and dose reduction. To recognize side effects as early as possible, it is of utmost importance to inform patients treated with alpelisib in detail about possible side effects and to monitor them closely.

In the recent trials as well as in our real-world setting, hyperglycemia is a common AE that occurs in almost two-thirds of all patients. So, before alpelisib therapy is initiated, patients should be checked carefully for preexisting glucose intolerance. Hyperglycemia due to alpelisib treatment should be managed by an interdisciplinary team. Based on our experience, we recommend including a diabetologist in the treatment team at the initiation of alpelisib therapy as well as equipping all patients with blood glucose monitors. According to the prescribing information and the information supplied by EMA, blood sugar levels should be monitored weekly, at least for the first 2 weeks, every 2 weeks until week 8, and then at least monthly [20, 21]. We had good experience with patients controlling their fasting blood sugar levels with the blood glucose monitor by themselves on a regular basis, in addition to the weekly lab controls at the Breast Center. HbA_1c should be controlled on day 1 and day 28 and at least every 3 months afterwards [20, 21]. In the case of AEs, more frequent controls are obviously necessary. Patients should be informed in detail about the symptoms of hyperglycemia (like polydipsia, polyuria, or increased appetite and weight loss) and ketoacidosis (additional nausea, vomiting, abdominal pain, acetone breath, and confusion) and should be instructed to consult a physician immediately they experience any symptoms. It is important to keep in mind that staging examinations via PET-CT are not possible in alpelisib-treated patients with hyperglycemia.

In our experience, rash was treated effectively with cortisone and antihistamines. Dermatologists should be consulted when rash occurs. In the BYLieve trial, a subgroup of 10 patients already received antihistamines before the occurrence of rash in cycle 1. This subgroup showed markedly reduced rates of rash compared to the subgroup without early use of antihistamines [15]. Therefore, prophylactic administration of oral antihistamines at the time of alpelisib therapy initiation may be considered. The recently updated ABC5 guideline contains the new statement “Patients receiving alpelisib in combination with endocrine therapy for PIK3CA-mutated ABC should be instructed to take non-sedating antihistamines to prevent rash at start of therapy. Antihistamines can be discontinued after 4 weeks as the risk for rash is primarily in the first 2 weeks of therapy” [6]. However, we had good experience with providing the patients with antihistamines to take as soon as any itching occurred, without administering antihistamines as a prevention in every patient. Patients should also be informed about optimizing skin care as described in the prescribing information [20, 21, 22].

Grade 3 diarrhea occurred in our small patient cohort more often than in the trials but could be treated effectively with supportive medical therapy (antidiarrheal drugs, fluids, and nutritional support), therapy interruption for short periods, and dose reduction. It is essential that patients are well-informed about this possible side effect and initiate symptomatic measures early. We recommend prescribing antidiarrheal drugs prophylactically to all patients and instruct them to start taking them as soon as symptoms occur.

Generally, it is essential to monitor all medications as well as relevant nonmedical therapies (e.g., phytotherapy) precisely, as interactions with alpelisib and contraindications must be considered. If unexpected side effects occur, they must be reported to the respective regulatory authorities.

All in all, when interdisciplinary supportive management is involved, alpelisib treatment has been shown to be feasible in clinical routine in metastatic breast cancer, in terms of both toxicity and efficacy. Although our patient cohort was more heavily pretreated than the trial cohorts, no patient stopped therapy due to toxicity and 1 patient was on the drug for >240 days.

Conclusion

The addition of the PI3K inhibitor alpelisib to ET with fulvestrant has become a new standard in HR+ HER2– PIK3CA-mutated ABC even if the optimal therapy sequence has not yet been fully elucidated. However, it is essential that tumor tissue (preferably from the most recent progression) of patients with HR+ HER2– ABC is tested for PIK3CA mutations, to evaluate the suitability for alpelisib treatment. Our real-world experience of almost 1 year demonstrates the necessity of patient education and an interdisciplinary approach when treating patients with alpelisib. Patients should be carefully monitored regarding possible therapy-associated side effects, especially hyperglycemia, rash, and diarrhea. An interdisciplinary team consisting of (gyneco-)oncologists, dermatologists, and diabetologists (from initiation of therapy onwards) should be included in patient care. Patients and their families as well as the responsible healthcare team (physicians, nurses, and the interdisciplinary team) must be trained in therapy management. The necessary information can be obtained from the prescribing information [20, 21, 23], specific guidelines [24], patient brochures, and a presentation on therapy management provided by the manufacturer. Training should include information about the possible side effects and their prophylaxis, the options of therapy interruption and dose reduction in the event of side effects, and drug interactions. Patient diaries or e-Health tools like CANKADO (www.cankado.com) can be very useful for monitoring this oral therapy.

Statement of Ethics

Since this is a retrospective report of our experience, there was no impact on the course of the therapies of individual patients. In consultation with the Ethics Committee, this project is considered a quality assurance project. All patients gave their written informed consent that the information from the MAP may be used for scientific purposes.

Conflict of Interest Statement

A.H. received honoraria for lectures and advisory boards from Roche and honoraria for lectures and reimbursement of training costs from Pfizer. N.H. received honoraria for lectures and/or consultancy work from Novartis and Roche. R.W. received honoraria for consultancy work as well as support for travel expenses from Agendia, Amgen, Aristo, AstraZeneca, Boehringer Ingelheim, Carl Zeiss, Celgene, Clinsol, Daiichi-Sankyo, Eisai, Genomic Health, Glaxo Smith Kline, Hexal, Lilly, Medstrom Medical, MSD, Mundipharma, Nanostring, Novartis, Odonate, Paxman, Palleos, Pfizer, Pierre Fabre, Puma Biotechnology, Riemser, Roche, Sandoz/Hexal, Seattle Genetics, Tesaro Bio, and Teva. All other authors declare no conflicts of interest.

Funding Sources

No funding was received.

Author Contributions

F.H. and R.W.: data collection. A.H., R.W., and N.H.: manuscript. A.H. and R.W.: analysis.

Supplementary Material

Supplementary data

verified