Prescribing and monitoring of riluzole in amyotrophic lateral sclerosis: a retrospective cohort study from Health New Zealand – Te Whatu Ora Waitaha Canterbury

Motor neurone disease (MND) is a progressive neurodegenerative disease affecting motor neurones of the brain and spinal cord. There are several clinical subtypes—the most common is amyotrophic lateral sclerosis (ALS), which affects both upper and lower motor neurons. Less common subtypes are defined by whether solely upper or lower motor neurones are affected (primary lateral sclerosis and progressive muscular atrophy respectively). The clinical presentation may include progressive weakness in the upper and lower limbs, problems with speech and swallowing and respiratory failure.¹ ALS is thought to affect around 350 people living in New Zealand at any one time.² There is no cure for MND, and care is focussed solely on optimising quality of life and support of patient functional goals.

Riluzole is the only funded life-prolonging treatment for the ALS type of MND in New Zealand.² Riluzole was initially approved by the United States of America Food and Drug Administration in 1995 after demonstrating a short survival benefit of 9% in 1-year survival (equating to 2–3 months overall).^1,3–4 No more effective treatments have yet been identified to slow, stop or reverse the course of ALS. Funding for treatment in New Zealand is based on specific eligibility criteria determined by Pharmac (see Table 1),⁵ and there are best-practice recommendations for monitoring as detailed by Motor Neurone Disease New Zealand.² Full blood count (FBC) and liver function tests (LFTs) are recommended at the baseline, monthly for 3 months, 3-monthly for 12 months and then annually thereafter. Baseline spirometry confirming forced vital capacity (FVC) is 60% of the predicted value is also recommended.²

View Table 1–3, Figure 1.

Objectives

The aim of this study was to assess adherence to the baseline and ongoing monitoring recommendations for MND patients in Canterbury as set out by Motor Neurone Disease New Zealand,² and by Pharmac in the Special Authority criteria. Secondary study measures included frequency of side effects to riluzole and reasons why riluzole was not prescribed.

Method

Canterbury is a large region in the centre of the South Island of New Zealand served by a single tertiary hospital based in Christchurch, where the Canterbury MND Service is based. The MND Service includes a multidisciplinary medical, nursing and allied health team, and all patients with a diagnosis of MND are referred to the service. Data collection involved retrospective review of the MND patient database maintained by the Christchurch Hospital respiratory department. Patients diagnosed between 1 January 2020 and 13 November 2024 were included in data collection.

Baseline demographic information was collected, including region and documented diagnosis. Patients were excluded if diagnosis was not a subtype of MND or records were incomplete, meaning prescribing of riluzole or starting date was unable to be determined (i.e., initial reviews in private health system, having moved from different region after onset of treatment).

Patient information was accessed via Health Connect South, a shared electronic patient record used throughout the South Island of New Zealand, alongside the online Special Authority records. Both neurology and respiratory letters were reviewed for documentation of riluzole prescribing and/or side effects, alongside confirmation with the online Special Authority system and Health Connect South medicines viewer. Spirometry and laboratory results were accessed via Éclair, an online patient database of laboratory test records.

Primary end points were based on adherence to monitoring requirements for patients on riluzole. These parameters were determined by Motor Neurone Disease New Zealand’s Best Practice Recommendations for the Care of People with Motor Neurone Disease.² The date of starting the medication was defined by the date of the Special Authority application and approval. At each monitoring time point, patients who had discontinued riluzole, died or not yet reached the monitoring time point were excluded from further analysis.

Secondary end points were time from diagnosis to prescription (defined by date of Special Authority application) and documented riluzole side effects or discontinuation.

Data summaries are shown as counts and proportions expressed as percentages and means and ranges for scale variables.

Ethics approval

The MND patient database is maintained for clinical use and not primarily for research purposes. As such, this study was performed as part of quality improvement processes and clinical governance and is exempt from national Health and Disability Ethics Committee requirements.

Results

Patient demographics and baseline characteristics

One hundred and nineteen National Health Index numbers were pulled from the MND database for patients diagnosed between 1 January 2020 and 13 November 2024. Four patients were excluded, with a total study cohort of 115 patients (see Figure 1, Table 2).

Patients were primarily based in Christchurch (81%), though this cohort also included a small number of patients based in South Canterbury, West Coast, Queenstown, Kaikōura and the Chatham Islands.

Seventy-nine of 115 patients had a documented initial diagnosis of ALS (68%), and 33 patients had a diagnosis of progressive bulbar palsy or MND with bulbar subtype (29%). There were three patients with an alternative diagnosis—these included paraneoplastic MND, MND with frontotemporal dementia and MND with Lewy body dementia (classified in Table 2 as “other”).

Of the total cohort, 69 patients (60.0%) had been prescribed riluzole. The average time from diagnosis to prescription of riluzole in this cohort was 1.05 months, ranging from 0 (i.e., at diagnosis) to 7 months following diagnosis. Table 2 reports the proportion of those prescribed riluzole within a certain time frame who had the recommended baseline testing completed. Fifty-nine percent of those prescribed riluzole were prescribed it at the time of diagnosis. Of those prescribed at diagnosis, 63% had baseline spirometry and just less than 50% had the complete baseline bloods. A further 26% of patients were prescribed riluzole within 3 months, and of this group nearly 90% had baseline spirometry though only 17% had baseline bloods. The remaining 15% of patients prescribed riluzole were prescribed more than 3 months following diagnosis. Sixty percent of this group had baseline spirometry and 20% had complete baseline bloods.

Of the patients who had been prescribed riluzole, the majority did not undertake the recommended baseline blood tests prior to treatment initiation. Over half of the cohort did not have baseline full blood count tested, and nearly 60% did not have baseline LFTs tested. Baseline spirometry was performed for 70% of patients prescribed riluzole.

The total number of participants who were not prescribed riluzole was 46 (37.9%). Of those, 20% had declined riluzole and 28% did not meet criteria for prescribing. Seventeen percent of those who were not prescribed riluzole would have been eligible based on baseline investigations, and 15% did not have the appropriate investigations to determine eligibility (see Figure 1).

Ongoing monitoring requirements

Overall, adherence to ongoing monitoring requirements was poor. In the initial phase of monitoring (13 months), 31% and 27% had the required tests at 1 and 3 months respectively. Both FBC and LFTs were performed for 58% of patients at 2 months, the highest proportion throughout the entire monitoring period spanning 1–36 months. Between 13% and 15% of patients only had LFTs monitored during the initial 3 months. Most patients did not undergo blood testing at 6, 9 and 12 months. Following this time point, sample groups were very small due to the number excluded.

Adverse reactions and cessation of riluzole

Eight patients (11.6% of those prescribed riluzole) experienced documented adverse events (AE). Three reported alanine aminotransferase (ALT) derangement greater than three times the upper limit of normal (30U/L), and three reported other AE (one each of constipation, incontinence and perioral dysaesthesia). Two had unspecified AE. Of these eight patients, seven ceased riluzole as a result. All discontinuations occurred within 9 months of starting the medication, and 63% (5 of 8) occurred within 3 months.

Discussion

This study has reviewed adherence to national best practice guidelines for the monitoring of patients on riluzole, as well as reasons for not prescribing or ceasing riluzole over a 4-year period in the Canterbury Region. It is promising that most patients in the cohort were prescribed riluzole, and that this was often prescribed promptly at diagnosis. However, this study highlights a significant disconnect between current clinical practice and the Motor Neurone Disease New Zealand best practice guidelines. More than half of the patients prescribed riluzole did not receive recommended baseline biochemical monitoring prior to starting. Of those prescribed riluzole at diagnosis, only 50% had baseline blood tests at that point. Delaying prescribing by up to 3 months did not improve monitoring rates—instead rates of biochemical monitoring declined from 50% to 15%. Over the longer-term monitoring period ranging from 1 month to 3 years, adherence to recommended blood test monitoring did not exceed 58%.

The MND best practice guidelines are concordant with the monitoring intervals suggested in the Cochrane review.³ The clinical recommendations for liver function monitoring are based on early studies demonstrating an increased incidence of elevated serum ALT levels in patients treated with riluzole compared with controls.^4,6 Haemoglobin monitoring stems from the observed trend towards lower haemoglobin in patients on riluzole in the 1996 study,⁴ though this result was not significant and there is no further evidence to support this. In this cohort, three patients ceased riluzole due to LFT derangement (ALT rise greater than three times the upper limit of normal), making up 4% of the total cohort prescribed riluzole. While the overall study numbers are small, this is a significant proportion of the treated group and emphasises the importance of regular biochemical monitoring. All three patients developed abnormalities within 9 months—two within 2 months of starting treatment. Initial prescribers of riluzole have a professional obligation to be aware of these recommendations and work proactively to fulfil them to deliver safe patient care. Acknowledging that the specialist (neurologist or respiratory physician) may not be the ongoing prescriber, the patient may benefit from a clear request in documentation to primary care colleagues to continue the monitoring pathway once the initial specialist contact is fulfilled. These requirements are clearly outlined by the current statement from the Medical Council of New Zealand | Te Kaunihera Rata o Aotearoa on good prescribing practice.⁷

We observed that spirometry was not completed at the time of application in approximately one-third of patients prescribed riluzole. However, all patients subsequently had spirometry performed and met criteria. It is important to recognise that while the formal criterion was not fulfilled at the point of application, prescribing was ultimately clinically appropriate and safe.

Special Authority approval criteria mirrored the trial populations from the two foundational randomised controlled trials (RCTs) in the 1990s, which demonstrated a modest survival benefit of 2–3 months in patients under 75 years of age, with disease duration of fewer than 5 years and FVC of greater than 60%.^4,6 There was no significant survival advantage observed in patients with more advanced disease (extrapolated by FVC <60%, tracheostomy or longer duration of symptoms beyond 5 years) or those over 75 years of age.⁸ While these criteria reflect the original evidence base, the necessity and relevance of these restrictions warrants critical examination. The evidence supporting the criteria is both limited and dated—no new RCTs have been published since, and these foundational studies are now nearly 30 years old. The FVC threshold of over 60% was originally intended to “improve the detection of outcomes”.⁶ Though pragmatic in the context of an RCT population, it is an arbitrary value that does not accurately reflect the complexity or variability of disease progression in actual clinical practice. Advances in diagnostics, multidisciplinary care and management of symptoms since the 1990s means that FVC may no longer be a necessary determinant of disease severity or have the potential to benefit from treatment.

Furthermore, emerging evidence from recent non-RCT studies indicates that the therapeutic benefit of riluzole may extend beyond initial estimates, both in terms of survival and the breadth of the patient population who may respond. A 2025 multicentre trial of 4,847 patients has demonstrated a 7-month longer median survival in patients treated with riluzole compared with those who did not receive riluzole (17.6 months compared with 10.7 months)⁹—significantly longer than the 2–3 months determined by the foundational studies.^4,6 Vasta et al. also demonstrated a survival advantage across the entire 10-year follow-up period, and noted patients classified as “fast progressors”—based on the rate of decline in ALS Functional Rating Scale scores—appeared to derive greater benefit than slower progressors.⁹ These findings align with those of a 2018 study, which reported that patients with lower FVC at trial entry experienced the most pronounced survival benefit from riluzole.¹⁰ In the Canterbury patient cohort, the most common reason documented for not prescribing riluzole was that the patient did not meet the eligibility criteria due to a sub-threshold FVC on spirometry or respiratory failure at the time of diagnosis. Taken together, these findings raise the possibility that current prescribing criteria may inadvertently exclude patients who stand to benefit the most—specifically, those with more advanced disease at the time of diagnosis.

The initial application for riluzole funding was submitted to Pharmac in June 2003, but approval was only granted in September 2013 following two previous rejections.^11–14 Initial rejections were cited to be due to “poor efficacy, high cost and poor cost-effectiveness”,¹² and despite a price reduction at the second application review the Pharmacology and Therapeutics Advisory Committee continued to state that “riluzole was still a very expensive treatment with little evidence of significant clinical benefit”.¹³ Riluzole was approved on its third submission, following a price reduction by the supplier.¹⁴ Although the price reduction enabled funding approval, riluzole remained accessible only through Special Authority due to its persistently high cost per quality-adjusted life year by Pharmac rapid analysis.¹⁴ In 2012, 1 year prior to approval, the cost of riluzole was cited as approximately GBP£4,000 or USD$10,000 annually.³ We do note that the Pharmac subsidy has reduced by fourfold since approval—the current cost of riluzole is approximately NZD$2 per tablet (NZD$117 per 56-tablet pack)¹⁵ compared with the NZD$7 per tablet (NZD$400 per 56-tablet pack) at the time of approval.¹⁶

When considered as a whole, the factors influencing the cost-effectiveness assessment or “cost per quality-adjusted life year” of riluzole have fundamentally changed. As a result, restrictive access to riluzole may no longer be justified, particularly when considering that this is the only life-prolonging therapy available.

This study has identified up to 37% of patients not prescribed riluzole who may have been eligible and benefitted from its use. This included eight patients who had undergone the required investigations and were assessed as eligible but never had a Special Authority application made, two who had an application approved but were never prescribed and seven who never had the required investigations completed. It is impossible, based on the retrospective data available, to determine if these patients are truly missed from the potential treatment cohort or if shared decision making and clinical rationale was applied appropriately.

Limitations of this study were predominantly related to the data collection method of a retrospective cohort study focussing on quantitative data. Information on side effects and reasons for not prescribing riluzole was sought from patient letters on Health Connect South; however, the letters were often not explicit. The date of prescribing was extrapolated from Special Authority application data for a similar reason—while letters may document if riluzole was discussed, it was unusual to see it clearly documented as prescribed on a particular date and, even so, dispensing data may not necessarily correlate with patient behaviour or date of starting treatment.

In summary, this study highlights inconsistencies between clinical practice and both the Motor Neurone Disease New Zealand best practice guidelines and Pharmac Special Authority criteria. Baseline and ongoing biochemical monitoring were often incomplete, despite known risks, and spirometry requirements were not consistently met at the time of Special Authority approval—though all patients were ultimately eligible. Current prescribing criteria are based on outdated evidence, with recent data suggesting greater and broader survival benefits from riluzole, especially in patients with more advanced disease. Future research should continue to explore the real-world outcomes of riluzole use in previously excluded patients, particularly those with advanced disease at diagnosis. Furthermore, updated cost-effectiveness analyses are necessary. As the evidence base expands, it is crucial that future policy decisions reflect this progress and support fair access to riluzole—the only treatment available to extend survival for MND patients in New Zealand.