A Phase I Dose-Escalation Study of Linsitinib (OSI-906), a Small-Molecule Dual Insulin-Like Growth Factor-1 Receptor/Insulin Receptor Kinase Inhibitor, in Combination with Irinotecan in Patients with Advanced Cancer.

LESSONS LEARNED: The maximum tolerated dose of the combination of linsitinib and irinotecan is linsitinib 450 mg daily on days 1-3 every 7 days and irinotecan 125 mg/m2 days 1 and 8 of a 21-day cycle.The adverse effects associated with the combination are not significantly increased beyond what is expected of each drug as a single agent.Multiple negative trials of insulin-like growth factor-1 receptor inhibitors performed in unselected patient populations led to the early discontinuation of linistinib development and this trial.Earlier integration of assessment of potential predictive biomarkers into clinical trials, as was planned in this study, is vital to the development of targeted therapies in oncology.
BACKGROUND: This phase I dose-escalation study was designed to evaluate the safety and tolerability of the combination of irinotecan and insulin-like growth factor-1 receptor (IGF-1R) inhibitor linsitinib in patients with advanced cancer refractory to standard therapy.
METHODS: Dose escalation in three specified dose levels was performed according to a standard 3 + 3 design. Dose levels were as follows: (a) linsitinib 400 mg and irinotecan 100 mg/m2, (b) linsitinib 450 mg and irinotecan 100 mg/m2, and (c) linsitinib 450 mg and irinotecan 125 mg/m2. Linisitinib was administered once daily on days 1-3, 8-10, and 15-17, and irinotecan on days 1 and 8. Assessment of a candidate predictive biomarker was planned in all patients, with further evaluation in an expansion cohort of advanced colorectal cancer.
RESULTS: A total of 17 patients were treated, with 1 patient in both cohort 2 and 3 experiencing dose-limiting toxicity. Linsitinib 450 mg and irinotecan 125 mg/m2 was the maximum tolerated dose. Sixteen (94%) patients experienced at least one treatment-related adverse event. Neutropenia was the only grade >3 toxicity (4%). No significant hyperglycemia or QT interval prolongation was noted. No objective responses were observed; 47% (n = 8) had stable disease with median duration of 5.25 months.
CONCLUSION: Although the combination was determined safe, the study was halted due to termination of linsitinib development, and biomarker testing was not performed. © AlphaMed Press; the data published online to support this summary are the property of the authors.

Discussion

Linsitinib is a potent small‐molecule tyrosine kinase inhibitor of the human IGF‐1R, with a half maximal inhibitory concentration (IC50) of 35 nmol/L, and the homologus insulin receptor, with an IC50 of 75 nmol/L. The drug is selective for these targets [1].

Irinotecan is a topoisomerase I inhibitor that is U.S. Food and Drug Administration approved for the treatment of colorectal cancer with compendia for reimbursement including non‐small cell lung, gastroesophageal, cervical, and ovarian cancers.

The combination of linsitinib and irinotecan was selected for further evaluation based on preclinical data suggesting a synergistic interaction between the drugs [2].

Eligible patients with refractory advanced cancer, and for which irinotecan is in the compendia for reimbursement, were treated with linsitinib, administered by mouth, and irinotecan, by intravenous (IV) infusion, in 21‐day cycles at three dose levels. Once the maximum tolerated dose (MTD) was defined, expansion of this dose level was planned in patients with advanced colorectal cancer. A potential predictive biomarker, the linsitinib integrated classifier score [3], was to be evaluated in this cohort.

A total of 18 patients were enrolled in the trial at a single site. One of seven evaluable patients in the second cohort experienced a dose‐limiting toxicity (DLT) of grade 3 nausea/vomiting requiring hospitalization. A DLT of grade 3 febrile neutropenia/grade 4 neutropenia was documented in one of seven patients treated in cohort 3. Linsitinib 450 mg and irinotecan 125 mg/m2 was determined to be the MTD.

The most common toxicities at least possibly related to treatment and occurring in at least 10% of cycles were nausea, vomiting, fatigue, and anorexia. Hyperglycemia and QTc prolongation were considered adverse events of special interest, although no events above grade 1 severity were documented.

Eight patients (47%) had stable disease. No responses were documented, although one patient with metastatic rectal cancer had a 23% decrease in tumor burden and was treated for 18 cycles. Seven patients (41%) had progressive disease.

Although the combination of linsitinib and irinotecan was determined to be safe at the MTD, the study was halted at this point due to termination of linsitinib development. Thus, the expansion cohort and analysis of the linsitinib integrated classifier and other pharmacodynamic and pharmacokinetic data were not completed.

Trial Information

Advanced colorectal, non‐small cell lung, gastroesophageal, cervical, and ovarian cancer

Metastatic/advanced

No designated number of regimens

Phase I

3 + 3

Maximum tolerated dose

Safety

Tolerability

Preliminary antitumor activity

Correlative endpoint

Drug tolerable, hints of efficacy

Drug Information

Linsitinib/OSI‐906

OSI Pharmaceuticals

Small molecule

Insulin‐like glistItemPairth factors—IGF‐1R and IGF‐2

mg per flat dose

p.o.

For cycle 1, patients were treated with a single dose of linsitinib on day −3, with further dosing days 2–4, 8–10, and 15–17. Patients received a single‐dose of linisitinb on days 1–3, 8–10, and 15–17 for all additional cycles.

Irinotecan

Camptosar

Pfizer

Other

Topoisomerase I

mg/m2

IV

Day 1 and 8 every 21 days for all treatment cycles.

Dose Escalation Table

Patient Characteristics

10

8

IV

Median (range): 51 (28–69)

Median (range): 2 (1–6)

0 — 9

1 — 9

2 — 0

3 — 0

Unknown — 0

Colon 10

Rectal 4

Esophageal 2

Cervical 1

Ovarian 1

Primary Assessment Method

Total patient population

21

18

17

12

RECIST 1.0

n = 0 (0%)

n = 0 (0%)

n = 8 (53%)

n = 7 (47%)

12 weeks

6 weeks

Best percentage change from baseline in sum of longest diameters

Adverse Events

Serious Adverse Events

Dose‐Limiting Toxicities

Assessment, Analysis, and Discussion

Study terminated before completion

Company stopped development

Drug tolerable, hints of efficacy

Although this study was discontinued early due to halting of linsitinib development, the dose‐escalation data do provide important safety information regarding this insulin‐like growth factor‐1 receptor (IGF‐1R) inhibitor in combination with irinotecan chemotherapy. In this study, the maximum tolerated dose of linsitinib was 450 mg daily on days 1–3 every 7 days in combination with irinotecan 125 mg/m2 days 1 and 8 of a 21‐day cycle. Overall, this combination was well tolerated across predefined dose levels, with most adverse events (AEs) grade 1–2 in severity.

Hyperglycemia is the primary class‐effect toxicity of IGF‐1R small‐molecule tyrosine kinase inhibitors (TKIs) due to insulin receptor (IR) cross‐targeting at clinically relevant doses [4], [5]. However, such AEs were overall mild in severity in this study, with no events meeting criteria for dose‐limiting toxicity (DLT) in this patient population. It is possible that no significant hyperglycemia was documented in this study because lower doses of linsitinib were used for combination dosing with irinotecan, and patients with baseline glucose elevations were excluded from participation. Elevation in liver function tests has also been documented in phase I studies of linsitinib alone and in combination with everolimus [5], [6], and although grade 3 elevation was observed in one patient on this trial, it was attributed to underlying disease and improved to grade 1 following stenting of a malignant stricture. Although not considered a class effect, QTc prolongation has been a DLT in other studies of linsitinib [4], [5], [7]. In this trial, no grade 3 or greater prolongation of QTc was observed. Unfortunately, due to early discontinuation of this clinical trial, we do not have pharmacokinetic data to further explore its relationship to this toxicity profile.

The early closure of this study and halting of linsitinib development is representative of the fate of IGF‐1R inhibitors in oncology drug development in the last 10 years. Although initially a promising target based on data from various preclinical studies, nearly 40 clinical trials evaluating IGF‐1R monoclonal antibodies, IGF‐1/2‐targeting antibodies, and IGF‐1R/IR small molecule TKIs did not demonstrate a significant clinical benefit in any tumor type [8], [9].

This includes studies evaluating IGF‐1R inhibitors in colorectal cancer, with both single‐agent trials [10] and combination studies with FOLFIRI [11], panitumumab [12], cetuximab/irinotecan [13], and everolimus [6] negative for a significant clinical benefit to patients. However, there were outlier patients across these studies who did achieve partial response or prolonged progression‐free survival on such therapy. It thus remains possible that a subset of colorectal cancer (CRC) patients may still benefit from IGF‐1R inhibitor therapy, although clearly a predictive biomarker is required to select such patients.

An important goal of the expansion cohort of this study was to explore this possibility in patients with advanced CRC; in this case using an integrated classifier to predict response to linsitinib therapy based on k‐Top Scoring Pair in combination with KRAS mutation status and IGF‐1R fluorescence in situ hybridization. Unfortunately, this attempt to identify a predictive biomarker for IGF‐1R targeted therapy came too late in the evaluation of this drug class, and the development of linsitinib was terminated before the classifier was explored in human patients.

Due to discontinuation of development of the majority of IGF‐1R inhibitors, there have been few other efforts to identify a biomarker predictive of activity within or across tumor types. However, a small number of ongoing clinical trials continue to evaluate this target in select tumor types thought to be dependent on IGF‐1R signaling, with the greatest interest in subtypes of sarcoma. Hopefully these and other ongoing studies specifically evaluating potential biomarkers of IGF‐1R inhibitor activity (NCT0271185, NCT02719041, NCT02916394) will lead to the identification of a predictive biomarker that will provide better identification of patients likely to benefit from IGF‐1R inhibition in the broader cancer patient population, as was an initial aim of this clinical trial.

All AEs in all cycles occurring in at least 10% of patients.Abbreviations: AGC, absolute granulocyte count; ANC, absolute neutrophil count; NC/NA, no change from baseline/no adverse event.

The five documented Serious Adverse Events occurred in five unique patients.