Kinetics and management of adverse events associated with lorlatinib after 5 years of follow-up in the CROWN study

Abstract

Objective

With 5 years of follow-up in the phase 3 CROWN study, lorlatinib showed unprecedented improvement in progression-free survival coupled with prolonged intracranial efficacy in patients with ALK-positive metastatic non-small cell lung cancer (mNSCLC). Here, we report kinetics and mitigation practices of select adverse events (AEs) to inform therapy management strategies.

Design

Post hoc safety analyses from the CROWN study assessed the incidence, prevalence, time to onset, duration, management, and resolution of hyperlipidemia, edema, weight gain, central nervous system (CNS) AEs, and peripheral neuropathy in the lorlatinib group (n = 149).

Results

After 5 years of follow-up, no new safety signals were observed. All-cause any-grade and grade 3/4 AEs occurred in 100% and 77% of patients, respectively; AEs led to lorlatinib dose reduction in 23% of patients, dose interruption in 62%, and permanent discontinuation in 11%. The median time to onset of any-grade hyperlipidemia was 0.5 months; 71% of events were managed with lipid-lowering agents. Median time to onset of any-grade edema, weight gain, CNS AEs, and peripheral neuropathy ranged from 2 to 4 months. Most weight gain events (95%) were mitigated with lifestyle modifications. Incidence and prevalence of CNS AEs did not increase over time; 58% of events did not require medical intervention.

Conclusions and Relevance

This post hoc analysis suggests that with longer lorlatinib exposure, no new safety signals emerged, and treatment discontinuation due to AEs remained low after 5 years of follow-up. Most AEs were effectively managed with dose modifications, indicating that current management strategies can be effective to mitigate toxicity. ClinicalTrials.gov NCT03052608

Graphical abstract

Graphical Abstract.

Implications for Practice.

Lorlatinib has shown unprecedented improvement in outcomes in treatment-naive patients with ALK-positive metastatic non-small cell lung cancer, with median progression-free survival (PFS) and median time to intracranial progression not reached after 5 years of follow-up. Given the durability of response, management approaches are critical for safely administering lorlatinib over prolonged periods. Dose modification was used effectively in the CROWN study resulting in low rates of discontinuation. Dose reduction did not impact PFS or intracranial efficacy. This manuscript summarizes the kinetics of adverse events (AEs) and management practices from CROWN to guide clinicians to effectively manage AEs associated with lorlatinib.

Introduction

Anaplastic lymphoma kinase (ALK) tyrosine kinase inhibitors (TKIs) are the standard first-line and subsequent therapy for patients with ALK-positive metastatic non-small cell lung cancer (mNSCLC).¹ Lorlatinib is a brain-penetrant, third-generation ALK TKI with broad-spectrum coverage of ALK resistance mutations.^2,3 Based on the results from a phase 1/2 study (NCT01970865) and the phase 3 CROWN study (NCT03052608), lorlatinib is approved for the treatment of patients with ALK-positive mNSCLC.^4–6 In the recent 5-year analysis from the CROWN study, lorlatinib continued to show progression-free survival (PFS) benefit and prolonged intracranial efficacy compared with crizotinib.⁷ With a median follow-up for PFS of 60.2 months for lorlatinib, median PFS was not reached (NR; 95% CI, 64.3 months-NR), corresponding to the longest PFS ever reported with any single-agent molecular targeted treatment in mNSCLC. At 5 years, the probability of being alive and free of progression was 60%. Median time to intracranial progression was also NR (95% CI, NR-NR) with lorlatinib, and the probability of being free from intracranial progression was 92% at 5 years.

Lorlatinib has a safety profile that is distinct from other ALK inhibitors and commonly observed toxicities include hypercholesterolemia, hypertriglyceridemia, edema, weight gain, central nervous system (CNS) adverse events (AEs), and peripheral neuropathy.^6–8 The safety profile of lorlatinib after long-term follow-up was consistent with those in earlier analyses,^6,8 with no new safety signals observed.⁷ At this 5-year analysis, treatment was ongoing in 50% of patients treated with lorlatinib; median duration of treatment was 57.0 months (IQR, 13.9-63.3 months). All-cause any-grade and grade 3/4 AEs occurred in 100% and 77% of patients, respectively. All-cause AEs led to dose reduction in 23% of patients, dose interruption in 62%, and permanent discontinuation in 11%; 5% of patients had treatment-related discontinuations, all of which occurred in the first 26 months.⁷ All other patients continued lorlatinib while deriving benefit with active management of toxicities, including dose interruptions or reductions, lifestyle modifications, and use of concomitant medications; some patients did not require any medical intervention.^9,10

Results from the CROWN study showed that with longer lorlatinib treatment, no new safety signals emerged, and rates of permanent lorlatinib discontinuation remained low after 5 years of follow-up. Active AE management throughout the study contributed to most patients continuing lorlatinib treatment. The present publication aims to further characterize the kinetics of select AEs and report the management practices followed in the CROWN study.

Methods

The CROWN study is an ongoing, international, open-label, randomized phase 3 trial comparing lorlatinib versus crizotinib in patients with previously untreated ALK-positive mNSCLC. Full details of the study design were published previously.^6–8 The protocol and amendments were approved by institutional review boards or independent ethics committees at each site and complied with International Guidelines for Biomedical Research Involving Human Subjects, Good Clinical Practice Guidelines, the principles of the Declaration of Helsinki, and local laws. All patients provided written informed consent.

Patients were randomized 1:1 to receive oral lorlatinib 100 mg once daily or crizotinib 250 mg twice daily. In this post hoc safety analysis, data from the 5-year follow-up of the CROWN study were used to assess time to onset, duration, incidence, prevalence, management, and resolution of select AEs in the 149 patients treated with lorlatinib.

The duration of an AE was defined as the time from the onset of the AE to its resolution or last available on treatment visit date for ongoing AE (duration = AE end date/last visit date − AE start date + 1 day). If a patient had multiple episodes of an AE, cumulative duration across all episodes was used adjusting for any overlap. For all AEs, resolved is reported as an outcome if the AE resolved or returned to the grade presenting at baseline. Partially resolved is reported as an outcome if the AE was not resolved but improved versus the worst grade observed. Not resolved was reported as an outcome if the AE does not fall in one of the two previous categories and for AE that led to permanent treatment discontinuation.

Results

After 5 years of follow-up and with longer exposure to lorlatinib in the CROWN study, no new safety signals were observed. The most common all-cause any-grade AEs leading to dose reduction (in ≥3% of patients) were edema (7%), hypertriglyceridemia (4%), cognitive effects (3%), mood effects (3%), and peripheral neuropathy (3%). In patients who had dose reduction (n = 49), median time to first dose reduction was 21.6 weeks (range, 2.1-281.9 weeks). The most common all-cause any-grade AEs leading to dose interruption (in ≥5% of patients) were hypertriglyceridemia (8%), pneumonia (8%), cognitive effects (6%), positive SARS-CoV-2 test (6%), edema (5%), peripheral neuropathy (5%), and mood effects (5%). AEs leading to permanent discontinuation (in >1% of patients) were cognitive effects (1%) and cardiac failure (1%). Post hoc safety analyses assessed hypercholesterolemia, hypertriglyceridemia, edema, weight gain, CNS AEs, and peripheral neuropathy.

Hyperlipidemia

Hypercholesterolemia and hypertriglyceridemia occurred in 108 (72%) and 99 (66%) patients, respectively (Figure 1). The median time to onset of any-grade hypercholesterolemia was 0.5 months (range, 0.03-39.6 months), and median duration was 35.9 months (range, 0.5-74.1 months; Figure 2). The median time to onset of grade ≥3 hypercholesterolemia was 6.4 months (range, 0.5-59.8 months), and median duration was 1.9 months (range, 0.03-32.5 months). The median time to onset of any-grade hypertriglyceridemia was 0.5 months (range, 0.2-32.0 months), and median duration was 37.8 months (range, 0.8-73.8 months). The median time to onset of grade ≥3 hypertriglyceridemia was 5.6 months (range, 0.3-56.1 months), and median duration was 2.6 months (range, 0.7-65.3 months).

Figure 1.

Summary of select adverse events. Abbreviation: CNS, central nervous system.

Figure 2.

Median time to onset and median duration of select adverse events. Duration = AE end date/last visit date for ongoing AE − AE start date + 1 day. If a patient had multiple episodes of an AE, cumulative duration across all episodes was used adjusting for any overlap. Abbreviation: CNS, central nervous system.

The majority (71%) of hyperlipidemia events were managed with lipid-lowering agents, 20% were managed without any medical intervention, and 5% were managed with lorlatinib dose interruption and lipid-lowering agents (Table 1). Overall, 45% of hyperlipidemia events resolved and 29% of events partially resolved; only 1 event led to permanent lorlatinib discontinuation. The most common concurrent medications for hyperlipidemia were rosuvastatin (55%), atorvastatin (19%), and fenofibrate (11%; Table S1).

Table 1.

Outcome of all-cause hyperlipidemia adverse events following lorlatinib dose modification and therapy management.

Intervention	Outcome, n (%)
	Resolved	Partially resolved	Not resolved	Total
Concurrent medication only	74 (31)	48 (20)	46 (19)	168 (71)
No medical intervention	24 (10)	6 (3)	17 (7)	47 (20)
Dose interruption + concurrent medication	4 (2)	9 (4)	0	13 (5)
Dose reduction + dose interruption + concurrent medication	2 (1)	4 (2)	0	6 (3)
Dose reduction + concurrent medication	2 (1)	1 (<1)	0	3 (1)
Permanent treatment discontinuation	0	0	1 (<1)	1 (<1)
Total	106 (45)	68 (29)	64 (27)	238 (100)

Edema and weight gain

Edema and weight gain occurred in 85 (57%) and 65 (44%) patients, respectively (Figure 1). Grade 1 weight gain occurred in 10 (7%), grade 2 in 21 (14%), and grade 3 in 34 (23%) patients. The median time to onset of any-grade edema was 1.8 months (range, 0.03-49.4 months), and median duration was 12.0 months (range, 0.1-67.2 months; Figure 2). The median time to onset of grade ≥3 edema was 4.4 months (range, 1.8-17.1 months), and median duration was 0.4 months (range, 0.3-15.9 months). The median time to onset of any-grade weight gain was 3.7 months (range, 0.03-38.7 months), and median duration was 27.4 months (range, 0.03-71.7 months). The median time to onset of grade ≥3 weight gain was 8.8 months (range, 1.9-62.5 months), and median duration was 22.7 months (range, 0.03-60.8 months).

To understand whether baseline body weight was associated with weight gain, baseline body weight was evaluated in patients with no weight gain or grade 1 weight gain (n = 94) and in patients who developed grade 2 or 3 weight gain (n = 55). The median baseline body weight was 64 kg (range, 32-105 kg) in patients who experienced no weight gain or grade 1 weight gain and 64 kg (range, 43-88 kg) in patients who developed grade 2 or 3 weight gain, suggesting that baseline body weight did not influence subsequent weight gain. In addition, a multivariate logistic regression analysis was employed to assess possible relationship between selected baseline characteristics and the development of weight gain; and the model only reported a reduction in risk of grade 2 or 3 weight gain associated with older age (odds ratio, 0.96; 95% CI, 0.93-0.99; Table S2). Edema and weight gain co-occurred in 40% of patients who reported edema and/or weight gain.

The incidence and prevalence of grade 2 and 3 weight gain combined did not increase over the 5-year period (Figure 3A). Ninety-five percent of grade 2 or 3 weight gain events did not require dose interruptions/modifications and/or concomitant medications (Table 2). Lorlatinib dose was modified in three patients (5%) whose weight gain did not resolve. Of the 52 patients who did not have a medical intervention for weight gain, 35%, 11%, and 49% experienced resolution, partial resolution, and no resolution of this AE, respectively.

Figure 3.

Incidence and prevalence of (A) grade 2/3 weight gain and (B) CNS effects. Abbreviation: CNS, central nervous system.

Table 2.

Outcome of all-cause grade 2 and 3 weight gain adverse events following lorlatinib dose modification and therapy management.

Intervention	Outcome, n (%)
	Resolved	Partially resolved	Not resolved	Total
No medical intervention	19 (35)	6 (11)	27 (49)	52 (95)
Dose reduction only	0	0	1 (2)	1 (2)
Dose interruption only	0	0	1 (2)	1 (2)
Dose reduction + dose interruption	0	0	1 (2)	1 (2)
Total	19 (35)	6 (11)	30 (55)	55 (100)

Central nervous system adverse events

CNS AEs occurred in 63 patients (42%), the majority of which (86%) were of grade 1 or 2 severity (Figure 1). Grade 1 CNS AEs occurred in 36 (24%), grade 2 in 18 (12%), grade 3 in 8 (5%), and grade 4 in 1 (1%) patient. As previously reported, cognitive effects occurred in 41 patients (28%), mood effects in 31 (21%), speech effects in 9 (6%), and psychotic effects in 8 (5%; Figure S1, see online supplementary material for a color version of this figure). The median time to onset of any-grade CNS AEs was 3.8 months (range, 0.03-54.4 months), and median duration was 7.8 months (range, 0.1-68.0 months; Figure 2). The median time to onset of grade ≥3 CNS AEs was 7.4 months (range, 1.1-53.1 months), and median duration was 0.7 months (range, 0.2-3.9 months). The incidence, prevalence, and severity of CNS AEs overall (Figure 3B) and by type (cognitive, mood, speech, and psychotic) did not increase over time (Figure S2, see online supplementary material for a color version of this figure). CNS AEs and weight gain co-occurred in 29% of patients who reported CNS AEs and/or weight gain.

After 5 years of follow-up, CNS AEs led to dose reduction in 8 (5%), dose interruption in 18 (12%), and permanent discontinuation in 3 (2%) patients (Table S3). Cognitive effects led to dose reduction in 4 (3%), dose interruption in 9 (6%), and permanent discontinuation in 2 (1%) patients. Mood effects led to dose reduction in 4 (3%) and dose interruption in 7 (5%) patients. Speech effects led to dose reduction in 1 (1%) and dose interruption in 4 (3%) patients. Psychotic effects led to dose reduction in 2 (1%), dose interruption in 2 (1%), and permanent discontinuation in 1 (1%) patient.

Of 118 CNS AEs, 58% were managed, per protocol, without any medical intervention, 14% were managed with dose interruption only, and 14% were managed with concurrent medication only (Table 3). Overall, 60% of CNS AEs resolved. Three events (confusional state, n = 2; delusion, n = 1) led to permanent lorlatinib discontinuation. The most common concurrent medications for CNS AEs were escitalopram, lorazepam, and quetiapine (2% each; Table S4).

Intervention	Outcome, n (%)
	Resolved	Partially resolved	Not resolved	Total
No medical intervention	39 (33)	1 (1)	28 (24)	68 (58)
Dose interruption only	14 (12)	1 (1)	1 (1)	16 (14)
Concurrent medication only	9 (8)	0	7 (6)	16 (14)
Dose reduction only	5 (4)	0	2 (2)	7 (6)
Dose interruption + concurrent medication	1 (1)	0	3 (3)	4 (3)
Dose reduction + dose interruption	2 (2)	1 (1)	0	3 (3)
Dose reduction + dose interruption + concurrent medication	1 (1)	0	0	1 (1)
Permanent treatment discontinuationa	0	0	3 (3)	3 (3)
Total	71 (60)	3 (3)	44 (37)	118 (100)

Abbreviation: CNS, central nervous system.

^aOccurred due to confusional state (n = 2) and delusion (n = 1).

Peripheral neuropathy

Peripheral neuropathy occurred in 65 patients (44%; Figure 1). The median time to onset of any-grade peripheral neuropathy was 3.7 months (range, 0.1-58.6 months), and median duration was 17.9 months (range, 0.2-65.4 months; Figure 2). The median time to onset of grade ≥3 peripheral neuropathy was 4.3 months (range, 3.0-5.6 months), and median duration was 3.1 months (range, 1.9-4.3 months).

Discussion

The safety profile of lorlatinib is well established based on results from the phase 1/2^3,5 and CROWN studies.^6–8 Based on results from the 5-year follow-up in the CROWN study, it is expected that the majority of patients will be treated with lorlatinib over prolonged periods of time. Given that safety and tolerability are critical for patients receiving long-term therapy, understanding the safety profile and management approaches is vital for continued treatment with minimal impact on quality of life. Several publications, consensus guidelines, and podcasts have provided detailed recommendations for managing AEs to guide clinicians, advanced practice providers, and oncology pharmacists.^10–16 A recently published therapy management article provided a pragmatic, simplified approach to lorlatinib treatment management that centers on 4 main principles: prepare the patient, monitor for AEs, manage AEs, and reassess.⁹ As treatment for ALK-positive mNSCLC has evolved toward increasingly potent and effective next-generation agents and earlier use of TKIs that can be administered over the long term, the “prepare, monitor, manage, reassess” paradigm is foundational to AE management and optimizing tolerability and quality of life. Importantly, these principles focus on the patient, assessing how burdensome any given AE is to the patient as a guide to lorlatinib management.

Based on preclinical data and pharmacokinetic studies, lorlatinib clinical studies used a 100-mg starting dose, the dose approved by regulatory bodies.^4,17 In vitro studies indicated that lorlatinib is a time-dependent inhibitor, as well as a net moderate inducer, of CYP3A.^17,18 Multiple days of oral lorlatinib at 100 mg/day induced CYP3A4 and resulted in autoinduction of lorlatinib metabolism, which stabilized by day 15. Age, sex, and body weight did not have a statistically significant effect on lorlatinib pharmacokinetics, and therefore, no dose adjustment is recommended based on these parameters.⁴ The exposure-related safety findings confirm that 100 mg/day is tolerable and support the approved guidance on dose modifications. Given that 62% of patients had a dose interruption, 23% had a dose reduction, and only 11% had permanent treatment discontinuation, the majority of patients on lorlatinib in the CROWN study were able to remain on therapy for as long as benefit was derived, highlighting the importance of prompt identification and management of AEs.⁷ With improved understanding of lorlatinib’s long-term safety profile, and to maximize tolerability over prolonged periods, clinicians should adopt a patient-centric approach with preemptive suggestions on the management of frequent and less frequent but potentially troublesome AEs and institute lorlatinib dose modification or lifestyle modifications at the earliest sign of toxicity. Prompt identification and management of low-grade AEs before they become more recalcitrant is important.

Several studies have shown that lorlatinib dose reductions do not adversely impact treatment outcomes in patients with ALK-positive mNSCLC.^19,20 In a retrospective real-world study, lorlatinib dose modification was an effective strategy to manage AEs without impacting efficacy.²¹ In this study, lorlatinib serum concentrations were higher in patients with grade ≥3 AEs than in those without AEs. Dose modifications reduced lorlatinib serum concentration and allowed for continuation of lorlatinib treatment. A post hoc analysis from the CROWN study assessed efficacy in patients who had lorlatinib dose reduction within the first 16 weeks and in those who did not.⁷ Dose reduction did not seem to impact PFS or time to intracranial progression. These studies indicate that lorlatinib dose modification is an important step to mitigate toxicity and continue long-term treatment with lorlatinib. It is important that patients who are fearful of reporting AEs to their healthcare provider be prepared and reassured that their clinician may need to adjust the lorlatinib dose but that this should not impact efficacy.

Consistent with findings in prior analyses, lorlatinib AEs generally occurred early after initiation of therapy.¹¹ In this analysis, most AEs occurred within the first 4 months of starting lorlatinib; grade ≥3 events occurred later, underscoring the importance of early identification and management. It is important to educate patients and their care partners about potential AEs and their onset. Better understanding of time to onset and resolution can help optimize management strategies.

Hyperlipidemia occurs in most patients and appears early, within 2 weeks of initiating lorlatinib. Pragmatic guidance suggests that patients should be monitored for elevated blood lipids 1 month after starting lorlatinib and then at each follow-up visit.⁹ Assessing lipid levels at baseline and frequently after initiating lorlatinib is key to proper intervention; grade ≥3 toxicity was brief, manageable, and only led to lorlatinib discontinuation in 1 case. Serum cholesterol and triglycerides should be monitored before initiating lorlatinib, at 1 and 2 months after initiation, and periodically thereafter.⁹ In the CROWN study, hyperlipidemia was primarily managed with concurrent medications such as statins, ezetimibe, fibrates, or PCSK9 inhibitors. Pitavastatin (2 mg orally once daily), pravastatin (40 mg orally once daily), and rosuvastatin (5-10 mg orally once daily for moderate-intensity or 20-40 mg orally once daily for high-intensity therapy) are the recommended statins based on their low involvement with specific CYP450 enzymes that can interact with lorlatinib (eg, CYP3A4) to avoid an increase in lorlatinib serum levels and elevated risk of AEs.^10,11,22 Per protocol, pitavastatin and rosuvastatin could be used during lorlatinib treatment without dose adjustment since CYP3A4 is not involved in their elimination. Pravastatin, fluvastatin, and atorvastatin should be used with caution during lorlatinib treatment, and dose adjustment of these statins may be necessary. Lovastatin and simvastatin are not recommended for use during lorlatinib treatment. For patients who are statin intolerant, PCSK9 inhibitors can be considered.²³ For severely elevated lipid levels, initiate lipid-lowering therapy and pause lorlatinib until total cholesterol is <400 mg/dL and triglycerides are <500 mg/dL and then resume lorlatinib at the same dose. Triglyceride levels of >1000 mg/dL are associated with acute pancreatitis, and management is based on its severity.⁹ Strategies include treatment with intravenous fluids, bowel rest, pain control, and admission to the intensive care unit. Long-term treatment focuses on lowering triglyceride levels with lipid-lowering agents. If hyperlipidemia is uncontrolled, resume lorlatinib at a reduced dose.⁹ Almost half (45%) of all hyperlipidemia events were resolved with successful management strategies and 29% were partially resolved.

With longer follow-up, there was no increase in cardiovascular AEs in patients treated with lorlatinib compared with those treated with crizotinib (28% in both groups).⁷ Although a higher number of patients in the lorlatinib versus crizotinib group had hyperlipidemia at baseline or during the study (134 vs 32, respectively), the incidence of cardiovascular AEs was lower with lorlatinib than crizotinib (28% vs 47%). These results suggest that hyperlipidemia associated with lorlatinib did not increase cardiovascular risk in these patients at 5 years.⁷ It is possible that this finding is associated with the close monitoring and management of hyperlipidemia per protocol in the CROWN study. It is, therefore, highly recommended that a similar approach should be followed in routine clinical practice when lorlatinib is used.

Weight gain developed over a period of months, with no clear correlation between baseline weight or edema and lorlatinib-associated weight gain. Edema and weight gain co-occurred in a fraction of patients, suggesting different pathogenic mechanisms for these AEs. Weight gain has also been observed in patients treated with other ALK TKIs, most notably alectinib.^24,25 The exact pathomechanism of ALK TKI–associated weight gain is not known; however, patients have reported an increase in appetite, leading to increased caloric intake, and thus, weight gain.^11,25 Patients and care partners should be counseled at lorlatinib initiation about the possibility of weight gain as an AE and that lorlatinib appears to stimulate appetite. It is difficult for patients to lose weight in the face of appetite stimulation. A proactive preventive approach with lifestyle modifications, such as food intake counseling, resources to mitigate hunger pangs, dietary advice, and exercise, is recommended.⁹ For patients in whom lifestyle modification does not appear to be sufficient, consider referral to a nutritionist or weight clinic for added patient support.¹¹ Most (95%) weight gain events in the CROWN study were managed without any medical intervention, as per protocol. Overall, 35% of all weight gain events resolved without any medical intervention; however, in the real world, a more proactive preventive approach is recommended. Currently there are limited data to guide the effective and safe use of GLP1 inhibitors in patients receiving ALK TKIs, although this remains an active area of interest and study.^25,26

A broad spectrum of CNS AEs can occur in patients receiving lorlatinib, mostly cognitive, mood, speech, and psychotic effects. These CNS AEs can be attributed to the ability of lorlatinib to penetrate the blood-brain barrier.^27,28 Results from the phase 1/2 study showed CNS AEs in 49% of patients, with cognitive effects in 29%, mood effects in 23%, speech effects in 11%, and psychotic effects in 9% of patients.²⁸ With long-term follow-up in the CROWN study, the incidence, prevalence, and severity of CNS AEs did not increase over time.⁷

There are no clear predictive factors for CNS AEs. No correlation was observed between occurrence of CNS AEs and lorlatinib-associated weight gain as these AEs co-occurred in only a fraction of patients. With long-term follow-up, CNS AEs occurred in 67% of patients with and 41% of patients without prior brain radiotherapy, indicating that the frequency of CNS AEs was higher in patients with prior brain radiotherapy; however, patient numbers are too small to draw definitive conclusions.^7,20 Another study showed statistically significantly higher rates of cognitive and speech effects in patients who had received CNS radiation than in those who had not.²⁸ It is possible that further disruption of the blood–brain barrier from CNS-specific therapies including radiation may increase the risk of developing CNS AEs.

Cognitive effects may include nervous system disorders such as memory impairment, cognitive disorder, amnesia, confusional state, attention disturbance, delirium, and mental impairment.^10,11 Patients have reported experiences such as “sluggish thought,” “fogginess,” “trouble connecting the dots,” difficulty multitasking, difficulty finding the right words, issues with short-term memory or recall, and confusion. Mood effects are reported as irritability, anxiety, depression, affect lability, personality change, and mood swings.^10,11 Most patients with mood effects describe feeling more irritable, more prone to impatience or anger, and sometimes more anxious. When patients first start treatment, some report feeling a little “energized” or even “buzzed.” Patients have also reported “feeling flat” or “feeling less excited about things.” Speech effects may include dysarthria, slow speech, and speech disorder.^10,11 Psychotic effects are mostly visual or auditory hallucinations. Giving clear examples of what type of CNS AEs patients may expect is important as the term “CNS adverse effects” will be insufficient for most.

The vast majority of CNS AEs were grade 1 or 2, and more than half (58%) of CNS AEs, per protocol, did not require any medical intervention. The most common medications used for CNS AEs, as previously reported, included anxiolytics, antidepressants, and antiepileptics.²⁰ It is imperative to prepare patients and care partners and to discuss the possible occurrence of these AEs so that they remain vigilant for symptoms; often care partners or individuals close to the patient, such as family members, will notice CNS AEs before the patient does, particularly when symptoms are subtle. Most CNS AEs are reversible with lorlatinib dose interruption or reduction, so early management and resolution are key. Nonpharmacological management strategies to minimize the impact of CNS AEs, such as setting reminders, mindfulness, meditation, and cognitive behavioral therapy, may be useful.⁹ To potentially manage other causes of neurocognitive and psychiatric impairment, visiting a specialist (eg, neurologist, psychiatrist) may be helpful. Overall, 60% of CNS AEs resolved. Grade ≥3 toxicity was brief with a shorter duration of 0.7 months, indicating that the majority of lorlatinib-associated CNS AEs will abate quickly. In an observational prospective study, neurocognitive evaluation showed that lorlatinib did not result in a sustained or significant decline in neurocognitive domains.²⁹ It is important that CNS AEs be quickly diagnosed, as early dose interruption can lead to rapid reversal of these AEs, with treatment resumption at the same or lower dose upon resolution, as clinically indicated.

Lorlatinib-associated peripheral neuropathy is typically mild and reversible following lorlatinib dose modifications.¹¹ Peripheral neuropathy is primarily reported as paresthesia, peripheral sensory neuropathy, and muscular weakness. Symptoms are commonly described as tingling, numbness, and pain in extremities at night. Lorlatinib may exacerbate symptoms in patients with preexisting carpal tunnel syndrome. Peripheral neuropathy associated with edema may respond to compression garments, raising the affected area above the heart, increasing exercise, limiting dietary salt, physiotherapy, and lymphedema drainage/massage.⁹ Pharmacological mitigation strategies such as treatments with vitamin B1 and vitamin B6 and medications (gabapentin or pregabalin) may provide symptom relief.

Conclusion

After 5 years of follow-up in the CROWN study, lorlatinib showed unprecedented efficacy in patients with ALK-positive mNSCLC.⁷ The systemic efficacy results coupled with protection from intracranial progression set a new benchmark for the treatment of ALK-positive mNSCLC. The updated analysis did not reveal any new safety findings, indicating that the duration of treatment is expected to be prolonged. It is of imperative importance that proper preparation, education, and vigilance is applied so that AEs with lorlatinib can be managed with effective strategies and dose modifications. In the CROWN study, most AEs were effectively managed with dose modifications, indicating that current management strategies are effective to mitigate toxicity without compromising the efficacy of lorlatinib. Knowledge and experience from the 5-year analysis in the CROWN study should help inform management practices in routine clinical practice when lorlatinib is used to achieve optimal outcomes for patients with ALK-positive mNSCLC.

Author contributions

Geoffrey Liu (Conceptualization, Methodology, Writing—original draft, Writing—review & editing), Benjamin J. Solomon (Conceptualization, Investigation, Writing—review & editing), Julien Mazieres (Conceptualization, Methodology, Writing—review & editing), Dong-Wan Kim (Conceptualization, Methodology, Writing—review & editing), Diego Cortinovis (Methodology, Writing—review & editing), Takako Inoue (Conceptualization, Methodology, Writing—review & editing), Richu Sharma (Conceptualization, Methodology, Writing—review & editing), Holger Thurm (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Writing—original draft, Writing—review & editing), Anna Polli (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing—original draft, Writing—review & editing), and Todd M. Bauer (Conceptualization, Methodology, Writing—original draft, Writing—review & editing)

Supplementary material

Supplementary material is available at The Oncologist online.

Funding

This study was sponsored by Pfizer.

Conflicts of interest

G.L. reports honoraria from AbbVie, Amgen, AstraZeneca, Bayer, Bristol Myers Squibb, Jazz Pharmaceuticals, Merck, Novartis, Pfizer, Roche Canada, and Takeda; consulting or advisory role at AstraZeneca/MedImmune, Novartis, Pfizer, Roche Canada, and Takeda; and participation on speakers bureau for AstraZeneca, Pfizer, and Takeda; and institutional research funding from AstraZeneca/MedImmune, Boehringer Ingelheim, Roche, and Takeda. B.J.S. reports honoraria from Amgen, AstraZeneca, Merck Sharp & Dohme, Pfizer, and Roche/Genentech; consulting or advisory role at Amgen, AstraZeneca, BeiGene, Bristol Myers Squibb, GlaxoSmithKline, Janssen, Lilly, Merck Sharp & Dohme, Pfizer, Roche/Genentech, and Takeda; institutional research funding from Sanofi; and patents, royalties, and other intellectual property from UpToDate. J.M. reports consulting or advisory role at AstraZeneca, Blueprint Medicines, Bristol Myers Squibb, Hengrui Therapeutics, Lilly/ImClone, Merck Sharp & Dohme, Novartis, Pfizer, Pierre Fabre, and Roche/Genentech; institutional research funding from AstraZeneca, Bristol Myers Squibb, Pierre Fabre, and Roche; and travel, accommodations, expenses from Bristol Myers Squibb, Pfizer, and Roche. D.-W.K. reports institutional research funding from Alpha Biopharma, Amgen, AstraZeneca/MedImmune, Boehringer Ingelheim, BridgeBio Pharma, Chong Kun Dang Pharmaceutical, Daiichi Sankyo, GlaxoSmithKline, Hanmi, IMBdx, inno. N, Janssen, Merus, Mirati Therapeutics, Merck, Merck Sharp & Dohme, Novartis, Ono Pharmaceutical, Pfizer, Roche/Genentech, Takeda, TP Therapeutics, Xcovery, and Yuhan. D.C. reports personal fees for advisory board from Amgen, AstraZeneca, Bristol Myers Squibb, Genzyme, Merck Sharp & Dohme, Novartis, Roche, Sanofi, and Seagen. T.I. reports honoraria from AstraZeneca, Bristol Myers Squibb, Chugai, Daiichi Sankyo, Merck Sharp & Dohme, Ono Pharmaceutical, and Takeda. R.S. reports no conflict of interest. H.T. is employed by and owns stocks in Pfizer. A.P. is employed by and owns stocks in Pfizer. T.M.B. reports consulting or advisory role at AVEO, Bayer, Lilly, Pfizer, and Sanofi; participation on speakers bureau for Bayer and Lilly; and travel, accommodations, expenses from Pfizer.

Data availability

Upon request and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual deidentified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.