Alectinib Dose Escalation Re-induces Central Nervous System Responses in ALK-Positive Non-Small Cell Lung Cancer (NSCLC) Patients Relapsing on Standard Dose Alectinib

Abstract

The central nervous system (CNS) is an important and increasingly recognized site of treatment failure in ALK-positive, non-small cell lung cancer (NSCLC) patients receiving ALK inhibitors. In this report, we describe two ALK-positive patients who experienced initial improvements in CNS metastases on standard-dose alectinib (600 mg twice daily), but subsequently recurred with symptomatic leptomeningeal metastases. Both patients were dose-escalated to alectinib 900 mg twice daily, resulting in repeat clinical and radiographic responses. Our results suggest that dose intensification of alectinib may be necessary to overcome incomplete ALK inhibition in the CNS and prolong the durability of responses in patients with CNS metastases, particularly those with leptomeningeal carcinomatosis.

Introduction

Central nervous system (CNS) metastases are frequent complications of anaplastic lymphoma kinase (ALK)-positive lung cancer, affecting nearly 30% of newly diagnosed metastatic patients.¹ Furthermore, among ALK-positive patients treated with the ALK inhibitor crizotinib, the CNS is a common site of relapse.² Among crizotinib-resistant patients enrolling on clinical trials of second-generation ALK inhibitors, the frequency of CNS metastases is approximately 60%.^3,4 Recently, it has also been observed that such relapses may include leptomeningeal and intramedullary metastases.⁵ Together, these complications are major causes of morbidity and mortality for ALK-positive NSCLC patients.

Alectinib is a second-generation ALK inhibitor that has demonstrated significant anti-tumor activity in ALK-positive NSCLC. In a phase I study conducted in the United States, alectinib was associated with an objective response rate (ORR) of 55% in patients previously treated with crizotinib.⁶ Similar anti-tumor activity was observed in a recent global phase II study of alectinib.³ Among 122 crizotinib-resistant, ALK-positive patients, alectinib produced an ORR of 50% and median progression-free survival of 8.9 months. Alectinib has also shown impressive activity in patients with CNS metastases, including intracranial objectives response rates of 42.9–52%.^3,6 Importantly, alectinib has also produced responses in patients with leptomeningeal disease^7,8 — a clinical feature that historically portends a dismal prognosis in NSCLC (median overall survival ~ 12 weeks).⁹ Based upon these encouraging early results, alectinib has been granted breakthrough therapy designation by the United States Food and Drug Administration for ALK-positive NSCLC patients previously treated with crizotinib.

Case 1

We previously reported the case of a 56 year-old man with metastatic, ALK-positive NSCLC complicated by leptomeningeal metastases.⁷ He received sequential treatment with crizotinib, the second-generation ALK inhibitor ceritinib, and whole brain radiation therapy (WBRT), but subsequently recurred with leptomeningeal metastases (Figure 1A). Thereafter, he began alectinib, achieving a rapid clinical and radiographic response (Figure 1B). Six months after starting alectinib, however, the patient developed recurrent word-finding difficulties, gait imbalance and progressive fatigue. Repeat neuroimaging revealed patchy, diffuse intramedullary enhancement in the thoracic spinal cord and enhancement along the surface of the cervical and thoracic cord, consistent with progressive leptomeningeal carcinomatosis (Figure 1C). Given the absence of alternative effective treatments, the patient was dose-escalated from 600 mg to 900 mg alectinib twice daily. Within several weeks, he experienced significant improvements in balance, cognition, and speech. A repeat spine MRI performed two months later showed interval improvement in intramedullary and leptomeningeal enhancement (Figure 1D). His systemic disease remained stable. From the point of initial dose escalation, the patient remained on alectinib 900 mg twice daily with continued response for a total of six months, at which time he was found to have asymptomatic progression in the leptomeninges. Alectinib was well tolerated at the higher dose (900 mg twice daily) with only grade 1 constipation noted.

Figure 1.

Sagittal T1-weighted post-gadolinium magnetic resonance images (MRI) of the thoracic spine of an ALK-positive patient treated with alectinib, depicting: A) leptomeningeal enhancement along the surface of the thoracic spinal cord (red arrow) and an intramedullary metastasis at the level of T11 (blue arrow) prior to alectinib treatment, B) interval improvement in leptomeningeal enhancement and resolution of an intramedullary T11 metastasis after 2 months of alectinib 600 mg twice daily, C) worsening intramedullary enhancement (red arrow) in the thoracic spinal cord after 6 months of alectinib (600 mg twice daily), and D) interval improvement in intramedullary enhancement in the thoracic spinal cord following dose escalation of alectinib to 900 mg twice daily.

Case 2

Case 2 involves a 34 year-old man with ALK-rearranged NSCLC complicated by brain and osseous metastases at the time of initial diagnosis. Following completion of WBRT and palliative RT to the right ilium, the patient began first-line crizotinib. He remained on crizotinib for approximately 2 years, during which time he underwent a craniotomy with metastatectomy, as well as palliative RT to several painful osseous metastases. In December 2013, he discontinued crizotinib and began carboplatin and pemetrexed. After two cycles of chemotherapy, treatment was interrupted for stereotactic radiosurgery (SRS) to four brain metastases.

In July 2014, the patient enrolled on a phase I/II trial of alectinib (NCT01871805). He responded systemically to a dose of 600 mg twice daily. In May 2015, however, the patient was hospitalized with headaches and confusion and was found to be in nonconvulsive, status epilepticus. A brain MRI at that time demonstrated interval development of new focal leptomeningeal enhancement at multiple sites (Figure 2A), consistent with CNS progression. The patient was started on corticosteroids and he ultimately required three anti-epileptic drugs to control his seizures. Upon discharge, he had improvement in symptoms; however, he remained with moderate cognitive dysfunction, including aphasia and short term memory loss. Following discussions with the study sponsor, the decision was made to dose-escalate alectinib to 900 mg twice daily. Within one month, the patient and his family reported significant improvements in headaches, cognition, word-finding ability and memory. The patient was able to taper his corticosteroids to a minimal dose. A repeat brain MRI after ~4 weeks at the escalated dose revealed near complete resolution of the leptomeningeal enhancing foci (Figure 2B). The patient has remained on alectinib 900 mg twice daily for 3.5 months with an ongoing response. In general, dose-escalated alectinib was well tolerated. Adverse events related to study drug included transient grade 3 hypophosphatemia and grade 1 fatigue, constipation and dyspepsia.

Figure 2.

Coronal T1-weighted post-gadolinium magnetic resonance images (MRI) of an ALK-positive patient, depicting: A) a focus of leptomeningeal enhancement in the left mesial temporal lobe (red arrow) that developed on alectinib 600 mg twice daily and B) near complete interval resolution of leptomeningeal enhancement after 4 weeks of alectinib 900 mg twice daily.

Discussion

Historically, dosing of targeted therapies in oncology has typically followed a one-size-fits-all approach, with dose reductions for toxicities. However, there is precedent for dose-escalation strategies following treatment failure. For example, in chronic myeloid leukemia (CML), imatinib dose escalation was a common treatment approach in patients progressing on therapy prior to the development of second-generation BCR-ABL inhibitors.¹⁰ Similarly, imatinib dose escalation has also been explored at the time of disease progression in patients with gastrointestinal stromal tumors (GISTs).¹¹ In NSCLC, “pulsatile” dosing of the EGFR inhibitor erlotinib has been investigated in EGFR-mutant patients with leptomeningeal metastases based upon concerns for inadequate CNS penetration with standard dosing.¹² This strategy, either alone or in combination with low-dose daily erlotinib, has demonstrated activity in EGFR-mutant patients with CNS metastases.^13,14

Like the experience with pulse-dose erlotinib, the two cases outlined above illustrate the potential importance of dose intensification of alectinib in ALK-positive patients with disease affecting certain sanctuary sites, like the CNS. In a phase I dose-finding study of alectinib that was conducted in the United States, a maximum tolerated dose was not identified.⁶ Ultimately, a recommended phase 2 dose (RP2D) of 600 mg twice daily was selected based upon combined safety, efficacy and pharmacokinetic data. However, doses up to 900 mg twice daily were also explored. Notably, alectinib 900 mg twice daily was associated with numerically higher peak concentrations and exposure levels over time compared to the RP2D. Specifically, at a dose of 900 mg twice daily, the mean (±SD) maximal plasma concentration (C_max) and area under the plasma concentration-time curve (AUC_0–10) were 1140 ± 448 ng/mL and 9840 ± 4620 ng•hr/mL, respectively. By contrast, at the RP2D of 600 mg twice daily, the mean C_max and AUC_0–10 were 676 ± 186 ng/mL and 5400 ± 1400 ng•hr/mL, respectively. Among 13 patients treated in the 900 mg twice daily cohort in this phase I trial, dose-limiting toxicities (DLTs) were reported in two patients. DLTs included grade 3 headache and grade 3 neutropenia requiring a dose delay of 7 days. Both patients remained on study, and each DLT resolved after dose reduction. Of note, among the two patients detailed in our report above, alectinib 900 mg twice daily was generally well-tolerated with no significant adverse events.

In animal models, alectinib produces relatively high brain-to-plasma ratios, ranging from 0.63–0.94.¹⁵ Clinically, measurable concentrations of alectinib have also been detected in cerebrospinal fluid (CSF) samples obtained from ALK-positive patients treated with alectinib.⁶ For example, in the report by Gadgeel and colleagues, five ALK-positive patients with CNS metastases had paired CSF and plasma samples available for analysis, revealing an apparent linear correlation between concentrations of free alectinib in the CSF and serum.⁶ Nonetheless, CNS concentrations of alectinib did not appear fully equivalent to systemic drug levels in the above studies. Our report reinforces these observations and suggests that inadequate ALK inhibition by alectinib may still underlie CNS disease progression in some patients.

Notably, despite a clinical and radiographic repeat response to dose-escalated alectinib in Case 1, the patient ultimately developed disease progression after six months. In addition to pharmacokinetic considerations, various molecular mechanisms of resistance to ALK inhibitors have been described. In particular, several ALK resistance mutations (e.g., G1202R, V1180L, and I1171 missense mutations) have been shown to confer resistance to alectinib.^16,17 In Case 1, however, the patient experienced CNS-only progression. We were thus unable to repeat a biopsy to evaluate for ALK resistance mutations or other molecular mechanisms of resistance. Moving forward, emerging technologies that allow for the detection of cell-free DNA in the circulation¹⁸ or CSF¹⁹ may provide insights into molecular mechanisms of resistance in such cases. In time, this may also inform therapeutic decision-making, including the relative roles of additional systemic therapy versus radiotherapy for CNS disease.

In summary, our findings suggest that dose intensification of alectinib may be helpful in prolonging the durability of responses in patients with CNS metastases, particularly those with leptomeningeal metastases. Further investigation of this strategy may be warranted.