Efficacy and tolerability of low versus standard daily doses of antiseizure medications in newly diagnosed focal epilepsy. A multicenter, randomized, single-blind, non-inferiority trial (STANDLOW)
The Collaborators (STANDLOW study group) presented in Appendix A.
Abstract
Objective
The STANDLOW trial investigated whether first-line antiseizure monotherapy with low doses has a similar efficacy to standard doses, but with fewer adverse events, improved quality of life, and reduced costs for the National Health System.
Methods
Multicenter, randomized, parallel-arm, single-blind, non-inferiority trial, comparing low dose versus standard dose of antiseizure medications (carbamazepine, levetiracetam, valproate, zonisamide, oxcarbazepine, topiramate, lamotrigine, gabapentin, lacosamide) in adults with newly diagnosed focal epilepsy.
Results
The intention-to-treat (ITT) population consisted of 58 randomized patients, 29 in the low dose arm and 29 in the standard dose arm, 27 (46.6%) females and 31 (53.4%) males, with an age between 18 and 87 years (median 54.9, IQR 32–71). The seizure type was focal impaired awareness seizures in 44 (75.9%) and focal aware seizures in 14 (24.1%). Etiology was unknown in 43 (74.1%) and structural in 15 (25.9%). At study entry, EEG was epileptiform in 28 (48.2%) and seizure frequency was low (≤2 seizures/month) in 41 (70.7%). The estimated relapse proportions at 12 months were 47% for the low dose and 48% for the standard dose, with a difference of 1% (95% CI: −30%; 27%). At the end of the study visit (12 months of follow-up, or immediately after seizure relapse or study withdrawal for other reasons, whichever came first), no differences in the number or severity of adverse events or quality of life measures were observed between the two treatment groups. The total drug-related costs over the entire study period were lower in the low dose arm (median per participant 253 € versus 475 € in the standard dose arm).
Significance
Although the efficacy of low doses versus standard doses appeared similar, non-inferiority could not be demonstrated due to slow recruitment and premature termination of the trial. Although statistically inconclusive, our findings suggest that a low dose of antiseizure medications may be considered as a first-line option in adult patients with a new diagnosis of focal epilepsy of unknown etiology and low seizure frequency.
Plain Language Summary
This study aimed to see if low doses of anti-seizure medications (ASMs) could be as effective as standard doses in treating adults with newly diagnosed epilepsy. Subjects were assigned to receive either a low or standard dose of ASMs. 58 adults participated. Both low and standard doses seemed to have a similar effect on controlling seizures. The study was stopped early due to slow enrollment, making it difficult to definitively prove that low doses were non-inferior to standard doses. Low doses of ASMs might be a reasonable option for adults with newly diagnosed epilepsy with no clear cause and few seizures.
Key points
- The estimated relapse proportions at 12 months were 47% for the low dose and 48% for the standard dose, with a difference of 1%.
- A low dose of antiseizure medications may be started in adults with a new diagnosis of focal epilepsy of unknown etiology and low seizure frequency.
- An equivalence of the low dose versus the standard dose of first-line ASM in the study sample is suggested.
1 INTRODUCTION
The efficacy of the first treatment in patients with newly diagnosed epilepsy is well known. About 33%–58% of patients achieve complete seizure remission with the first maintenance daily dose, defined as the dosage on which a patient is stabilized after the initial titration phase.1 However, with few exceptions, individual trials and Cochrane reviews show similar efficacy with differing maintenance dosages.2 The initial maintenance dosage is rarely discussed in published reports, and the lack of specifically designed studies means that no consensus exists about what dosage should be used.
Published reports and common experience show that exposing patients to long-term treatment with higher dosages of antiseizure medications (ASMs) increases the risk of adverse effects and has a negative impact on quality of life.3 On the contrary, when ASM treatment is given at low doses, the frequency and type of adverse events are not different from untreated individuals.4 There are no published studies comparing low to standard doses of ASMs.
In 1996 the promoter started a nationwide randomized pragmatic trial comparing low to standard doses of carbamazepine, phenobarbital, phenytoin, and valproate in children and adults with newly diagnosed epilepsy,5 but small sample size prevented providing definite conclusions.
On this background, the aim of the STANDLOW (STANDard versus LOW daily doses of antiseizure medications) trial was to test whether low doses of ASMs in adults with newly diagnosed focal epilepsy were at least as effective as standard doses of ASMs and carried a lower risk of adverse effects and translated into a better quality of life and satisfaction, with reduced costs for the National Health System.
2 METHODS
This was a multicentre, randomized, pragmatic, parallel-group, single-blind, non-inferiority trial conducted in 12 Epilepsy Centers in Italy. Each center has been selected on the basis of high standards in the management of epilepsy and previous participation in multicenter clinical studies.
Patients were eligible for inclusion according to the following criteria: age 18 years or older; newly diagnosed and previously untreated epilepsy, defined according to the ILAE6; having experienced focal and/or focal-to-bilateral seizures, defined according to the ILAE criteria7; able to understand and comply with the study requirements and to sign a written informed consent.
Exclusion criteria were: age under 18 years; presence of primary generalized tonic–clonic seizures, or other types of non-focal seizures; previous exposure to ASMs; inability to understand the aims and methods of the study; current pregnancy or pregnancy planning during the study period; inability to use contraceptive methods for the entire duration of the study; need to use a standard or low dose of the drug for specific needs; poor compliance; refusal to sign the informed consent.
The treating Epileptologist (treating physician) was free to choose the ASM according to the patient's clinical needs and his/her own clinical judgment, while the dose of the drug was randomly determined. The physician evaluating outcomes (evaluating physician; see below) was aware of the type and blinded to the dose of ASM. The treating physician, as well as the patient, was aware of both the type and the dose of ASM.
After informed consent, eligible patients were randomized to receive a low or a standard dose of the chosen drug. Randomization was centralized at the Istituto di Ricerche Farmacologiche Mario Negri IRCCS, Milano. The randomization sequence was created using SAS 9.4 (SAS Institute) statistical software and was stratified by center with a 1:1 allocation using random block sizes of 4. Each center received randomization scratch cards.
The drugs under study were the ASMs available for monotherapy treatment in Italy at the time of study recruitment: carbamazepine, levetiracetam, valproate, zonisamide, oxcarbazepine, topiramate, lamotrigine, gabapentin, and lacosamide.
Eligible subjects were randomized to receive low or standard doses of the chosen drug (carbamazepine, 300 mg/day versus 600 mg/day; levetiracetam, 500 mg/day versus 1000 mg/day; valproate, 300 mg/day versus 600 mg/day; zonisamide,150 mg/day versus 300 mg/day; oxcarbazepine, 450 mg/day versus 900 mg/day; topiramate, 100 mg/day versus 200 mg/day; lamotrigine, 100 mg/day versus 200 mg/day; lacosamide 100 mg/day versus 200 mg/day; gabapentin, 450 mg/day versus 900 mg/day).
The recruitment period lasted 27 months and patients enrolled at the baseline visit were followed up for 12 months from randomization or until the first seizure relapse, or until study withdrawal for other reasons, whichever came first. In the event of a known or suspected seizure, or of a serious adverse event, patients were instructed to call a member of the practice staff within 24 h. If the study staff was certain of the epileptic nature of the event, the patient terminated the study and an end-of-study visit was arranged within 72 h.
Follow-up visits were organized at the discretion of the physician, while phone contacts were scheduled at 4 weeks, 3 months, 6 months, and 9 months. An end-of-study visit was arranged at 12 months or immediately after the first seizure relapse, or until study withdrawal for other reasons, whichever came first. All visits were carried out within an interval of ±1 week compared to the expected exact timing. During phone contacts, the patient was asked to communicate any recurrence of seizures, treatment side effects, adverse events, and any other information deemed important for the management of the disease.
2.1 Data collection at baseline and during follow-up
The following variables were collected at admission by the evaluating physician: date of birth, sex, height, weight, arterial blood pressure, heart rate, family history of epilepsy, history of febrile seizures, date of onset of seizures, seizure type(s), epilepsy syndrome or non-syndromic epilepsy, neurological and psychiatric examination (coded as normal or abnormal), first interictal EEG (coded as normal, slow or epileptiform), imaging findings (coded as normal or abnormal for seizure etiology), relevant comorbidities, concurrent treatments, HRQOL (using the Italian version of the QOLIE-31 inventory8), patient satisfaction (PSQ-18), laboratory examinations if available (blood count, alanine aminotransferase, aspartate aminotransferase, gamma glutamyltranspeptidase, creatinine, alkaline phosphatase, calcium, phosphorus, sodium, potassium, ammonia, amylase). Each patient was followed for 12 months. Laboratory tests were performed as per clinical practice or caring physician discretion, including the ASM plasma levels where required by clinical practice. Ad-hoc diaries were given to each patient and/or caregiver for the collection of follow-up data. The diaries included the number of seizures (with dates) and the type and severity of adverse events (with dates, outcome and actions). QOLIE-31 and PSQ-18 questionnaires were administered a second time at the end-of-study visit (at 12 months or earlier, in case of seizure relapse). To preserve blindness, daily intake and timing of the assigned drug, but not the daily dose, were recorded in the diaries. Any contact with the National Health System for the management of seizures and ASMs was also noted in the diaries with dates. The content of the diaries was verified during each phone contact by the evaluating physician and by the treating physician at the time of the face-to-face visits.
2.2 Study outcomes
Outcomes were evaluated by physicians aware of the type and blinded to the dose of ASM (evaluating physicians).
The primary outcome of the study was the proportion of patients who experienced ASM failure, defined as a seizure relapse of any type.
- The proportion of patients who experienced treatment failure caused by intolerable adverse events related to the ASMs, resulting in the need to change the dose or drug.
- The average score of the QOLIE-31 scale, which measures the quality of life in patients with epilepsy (range 0–100, where 0 indicates the worst quality of life, 100 the best), at the end-of-study visit;
- The average of the scores relating to the 7 domains of the PSQ-18 scale, which measure various aspects related to satisfaction in disease management (general satisfaction, technical quality, interpersonal manner, communication, financial aspects, time spent with doctors, accessibility and convenience), each with a range from 1 to 5, where 1 indicates the least possible satisfaction and 5 the greatest, at the end-of-study visit;
- The median costs per patient and total costs due to the management of epilepsy during the 12 months of the study.
2.3 Statistical analysis
The statistical analysis plan of the study included descriptive statistics for the main demographic and clinical characteristics of the randomized patients, divided by treatment arm (low dose and standard dose). Continuous variables were described by means, with standard deviation or medians with interquartile range or range, while the categorical ones were described by frequencies and percentages.
The proportion of patients experiencing the primary outcome was estimated using Kaplan–Meier survival curves over the entire (12-month) follow-up period. The difference between the proportions of patients who relapse, with a 95% confidence interval (95% CI), in the 2 treatment arms (low dose-standard dose) was calculated at 3, 6, 9, and 12 months. This difference, with its 95% CI, was then compared with the chosen non-inferiority margin (m = 0.15). Non-inferiority was declared if the confidence interval for the difference between the proportions for the low dose versus standard dose was entirely below the non-inferiority margin. In case of imbalances between the 2 treatment arms with respect to variables that could have an effect on the outcome (potential confounders), a multivariate analysis was planned, using a Cox model adjusted for these variables.
Treatment failure due to adverse events (secondary endpoint) was analyzed using Kaplan–Meier survival curves, comparing the two treatment arms with the log-rank test. In the presence of possible confounders, the use of a multivariable Cox model, adjusted for these confounders, was planned.
The total number of adverse events, severe adverse events, and serious adverse events over the entire trial duration was compared between the two treatment arms with the Wilcoxon-Mann–Whitney test.
The median scores (and interquartile range, IQR) for the QOLIE-31 scale and for the domains of the PSQ-18 scale were calculated in the two treatment arms at the end of the study visit. Scores were compared between arms using the Wilcoxon-Mann–Whitney test. In addition, the median (and IQR) changes between baseline and the end of the study visit for both the QOLIE-31 scale and the 7 domains of the PSQ-18 were calculated and compared between the two treatment arms with the Wilcoxon-Mann–Whitney test.
The primary analysis was performed in the ITT population, including all randomized patients. A secondary analysis was repeated in the per-protocol (PP) population, after excluding patients with significant protocol deviations that could have an impact on the outcome (unverified compliance, follow-up visits performed more than 1 month before or after the timing planned by protocol).
Finally, subgroup analyses were performed by sex, age group, type of seizure, etiology, type of drug used, and center.
2.4 Sample size calculation
The sample size was based on the primary efficacy end-point. A relative 20% difference, up to a maximum of an absolute 15% difference in seizure relapse could be considered as non-inferiority. The non-inferiority margin for the difference between the two treatment arms in the proportion of seizure relapse was fixed at 15%. Assuming that the proportion of relapses in the low and standard dose groups is 60%, 168 patients per group (336 total) provide 80% power to detect non-inferiority with a one-sided 2.5% level of significance (Farrington-Manning test). Assuming a10% drop-out rate, the required sample size was 374 patients.
3 RESULTS
The trial was terminated before reaching the predefined sample size due to slow recruitment and exhaustion of time for enrolment. Between March 2021 and June 2023, 103 patients were screened, and 58 patients were randomized in 12 participating centres. The reasons for exclusion were the following: diagnosis other than epilepsy (n = 9); need to use a standard or low dose of drug for specific needs (n = 29); patients already being treated at screening (already treated with an ASM prescribed by the general practitioner or emergency room doctor) (n = 2); primarily generalized seizures (n = 7).
The total number of randomized patients, dropouts, and protocol deviations, in total and by treatment arm, is shown in Table S1.
3.1 ITT analysis
The ITT population consisted of all 58 randomized patients, 29 in the low-dose arm and 29 in the standard-dose arm. The demographic and clinical characteristics of the randomized patients by treatment arm are shown in Table 1. Details of the ASM administered and their respective doses are shown in Table 2.
| Low dose | Standard dose | |||
|---|---|---|---|---|
| n/median | %/IQR | n/median | %/IQR | |
| Sex | ||||
| Male | 14 | 48.3 | 17 | 58.6 |
| Female | 15 | 51.7 | 12 | 41.4 |
| Age | 51.6 | 32.0–72.7 | 55.5 | 35.3–71.2 |
| Years of education | 13 | 8–13 | 12 | 8–13 |
| Working position | ||||
| Employed | 17 | 58.6 | 11 | 39.3 |
| Student | 8 | 13.8 | 2 | 7.1 |
| Retired | 4 | 27.6 | 10 | 35.7 |
| Unemployed | 0 | 0.0 | 5 | 17.9 |
| ND | 0 | – | 1 | – |
| Marital status | ||||
| Single | 12 | 41.4 | 12 | 41.4 |
| Married/living with partner | 15 | 51.7 | 15 | 51.7 |
| Separated | 0 | 0.0 | 0 | 0.0 |
| Divorced | 1 | 3.5 | 0 | 0.0 |
| Widowed | 1 | 3.5 | 2 | 6.9 |
| Family history of epilepsy | 10 | 35.7 | 7 | 24.1 |
| Family history of febrile seizures | 1 | 3.6 | 2 | 6.9 |
| Time from seizure onset to diagnosis (months) | 5.7 | 0.4–28.0 | 3.2 | 0.1–21.5 |
| Type of focal seizure | ||||
| Aware | 10 | 34.5 | 4 | 13.8 |
| Impaired awareness | 19 | 65.5 | 25 | 86.2 |
| Motor onset | 0 | 0.0 | 0 | 0.0 |
| Non motor onset | 0 | 0.0 | 0 | 0.0 |
| Focal to bilateral tonic–clonic | 0 | 0.0 | 0 | 0.0 |
| Seizure frequency | ||||
| <2 per month | 20 | 69.0 | 21 | 75.0 |
| 2–5 per month | 4 | 13.8 | 4 | 14.3 |
| 6–10 per month | 2 | 6.9 | 1 | 3.6 |
| >10 per month | 3 | 10.3 | 2 | 7.1 |
| ND | 0 | – | 1 | – |
| Etiology | ||||
| Structural | 5 | 17.2 | 10 | 34.5 |
| Genetics | 0 | 0.0 | 0 | 0.0 |
| Infectious | 0 | 0.0 | 0 | 0.0 |
| Metabolic | 0 | 0.0 | 0 | 0.0 |
| Immune | 0 | 0.0 | 0 | 0.0 |
| Unknown | 24 | 82.8 | 19 | 65.5 |
| QOLIE-31 | 69 | 50–80 | 60 | 46–65 |
| PSQ-18 | ||||
| General satisfaction | 5 | 4–5 | 4 | 4–5 |
| Technical quality | 4 | 4–5 | 4 | 4–5 |
| Interpersonal manner | 4 | 4–5 | 4 | 4–5 |
| Communication | 4 | 4–5 | 5 | 4–5 |
| Financial aspect | 4 | 3–5 | 4 | 4–5 |
| Time spent with doctors | 4 | 3–4 | 4 | 3–4 |
| Accessibility and convenience | 4 | 4–5 | 4 | 3–5 |
| First EEG | ||||
| Normal | 4 | 13.8 | 8 | 27.6 |
| Aspecific | 3 | 10.3 | 1 | 3.5 |
| Slow | 9 | 31.0 | 5 | 17.2 |
| Epileptiform | 13 | 44.8 | 15 | 51.7 |
| Imaging | ||||
| Normal | 17 | 58.6 | 15 | 51.7 |
| Abnormal | 11 | 37.9 | 14 | 48.3 |
| Not performed | 1 | 3.5 | 0 | 0.0 |
| ASM | Low dose | Standard dose | ||
|---|---|---|---|---|
| Dose mg/die | N | Dose mg/die | N | |
| Carbamazepine | 300 | 6 | 600 | 3 |
| Lacosamide | 100 | 5 | 200 | 9 |
| Lamotrigine | 100 | 3 | 200 | 3 |
| Levetiracetam | 500 | 13 | 1000 | 10 |
| Oxcarbazepine | 450 | 2 | 900 | 2 |
| Valproate | 300 | 0 | 600 | 2 |
During the 12 months of the study, a total of 11 patients treated with the low dose and 11 treated with the high dose experienced a treatment failure due to seizure relapse, while 3 patients in each treatment arm failed therapy due to intolerable adverse events. In addition, 2 patients treated with the low dose and 4 patients treated with the standard dose dropped out of the study for other reasons.
The probabilities (estimated proportions) of relapse during the 12 months of the study in the two treatment arms are shown in Figure 1. At month 12, the cumulative probability of relapse was 47% in the low dose arm and 48% in the standard dose arm. The difference between the estimated proportions for the low dose versus the standard dose was 1% (95% CI: −30%; 27%). Since the confidence intervals included the non-inferiority margin (15%), results were inconclusive and it was not possible to declare non-inferiority.
The estimated proportions of treatment failure (suspension or dose change) due to intolerable adverse events during the 12 months of the study in the two treatment arms are shown in Figure 2. No differences were detected between the two treatment arms (Log-rank p = 0.6560).
Even considering the total number of adverse events, severe adverse events, and serious adverse events over the entire trial duration, no differences were found between the two treatment arms (Table 3).
| Low dose | Standard dose | p-value | |||||
|---|---|---|---|---|---|---|---|
| Minimuma | Maximuma | Total | Minimuma | Maximuma | Total | ||
| Adverse events | 0 | 4 | 22 | 0 | 4 | 27 | 0.8728 |
| Severe adverse events | 0 | 2 | 7 | 0 | 2 | 10 | 0.5275 |
| Serious adverse events (SAE) | 0 | 2 | 3 | 0 | 1 | 1 | 0.5591 |
- a Per subject.
No differences were detected in the quality of life (QOLIE-31), nor in satisfaction with care (PSQ-18) at the end of the study visit (Table 4).
| Low dose | Standard dose | p-value | |||||
|---|---|---|---|---|---|---|---|
| Median | Lower quartile | Upper quartile | Median | Lower quartile | Upper quartile | ||
| QOLIE-31 at EOS | 68 | 56.5 | 76.5 | 64.5 | 55 | 78 | 0.7494 |
| PSQ-18 at EOS | |||||||
| General satisfaction | 4.5 | 3 | 5 | 4 | 3 | 4 | 0.3724 |
| Technical quality | 4 | 3.5 | 5 | 4 | 4 | 4 | 0.2761 |
| Interpersonal manner | 5 | 4 | 5 | 4 | 3 | 5 | 0.0913 |
| Communication | 5 | 3.5 | 5 | 4 | 4 | 5 | 0.7549 |
| Financial aspect | 4.5 | 4 | 5 | 4 | 4 | 5 | 0.5748 |
| Time spent with doctors | 4 | 3.5 | 4 | 4 | 3 | 4 | 0.5053 |
| Accessibility and convenience | 4 | 3.5 | 5 | 4 | 4 | 4 | 0.4274 |
| QOLIE-31 difference between EOS and baseline | 0 | −7 | 6 | 7 | −4 | 13 | 0.2247 |
| PSQ−18 difference between EOS and baseline | |||||||
| General satisfaction | 0 | −1 | 0 | 0 | −1 | 0 | 0.4293 |
| Technical quality | 0 | −1 | 1 | 0 | −1 | 0 | 0.2042 |
| Interpersonal manner | 0 | −1 | 1 | −0.5 | −1 | 0 | 0.1109 |
| Communication | 0 | −1 | 1 | 0 | −1 | 0 | 0.5984 |
| Financial aspect | 0 | −1 | 1 | 0 | −1 | 1 | 0.4298 |
| Time spent with doctors | 0 | −1 | 1 | 0 | −0.5 | 0.5 | 0.9028 |
| Accessibility and convenience | 0 | −1 | 1 | 0 | −0.5 | 1 | 0.8764 |
- Abbreviation: EOS, end of study visit.
Total drug-related costs over the entire study period were 6247.06 € (median per participant 252.46; IQR 167.90–252.46; range 54.75–338.93) in the low dose arm and 13 251.93 € (median per participant 474.50; IQR 338.92–677.86; range 109.50–677.86) in the standard dose arm.
3.2 Per-protocol analysis
The per-protocol population consisted of 48 patients, 24 randomized to receive the low dose and 24 randomized to receive the standard dose.
During the 12 months of the study, a total of 10 patients treated with the low dose and 10 treated with the standard dose failed therapy (changed dose or drug) due to a relapse, while 3 patients in each therapeutic arm failed therapy for intolerable adverse events. One patient treated with the low dose (poor compliance) and one patient treated with the standard dose (withdrawal of informed consent) left the study for other reasons. The probabilities (estimated proportions) of relapse during the 12 months of the study in the two treatment arms are shown in Figure S1. At month 12, the cumulative probability of relapse was 51% in the low dose arm and 49% in the standard dose arm. The difference between the proportions was 2% (95% CI −29%; 32%). The confidence interval included the non-inferiority margin (15%) and it was therefore not possible to declare non-inferiority.
No differences were detected regarding the probability of treatment failure (withdrawal or dose change) due to intolerable adverse events (Figure S2) (Log-rank p = 0.6765).
Considering the total number of adverse events, severe adverse events, and serious adverse events over the entire trial duration, no differences were found between the two treatment arms (Table S2).
No differences were detected in the quality of life (QOLIE-31), nor in satisfaction with care (PSQ-18) at the end of the study visit (Table S3).
Total drug-related costs over the entire study period were 4923.07 € (median per participant 252.46; IQR 167.90–252.46; range 54.75–338.93) in the low dose arm and 10 611.72.93 € (median per participant 474.50; IQR 337.36–677.86; range 109.50–677.86) in the standard dose arm.
3.3 Subgroup analyses
No differences between treatment arms were detected in any of the subgroup analyses, that were severely underpowered (data not shown).
4 DISCUSSION
The STANDLOW trial could not reach the target sample size due to slow recruitment and exhaustion of time for enrollment. Two major reasons contributed to slow recruitment: first, the COVID pandemic slowed or halted the activation procedures of many centers and hampered the participation of both clinical investigators (who were busy with the extraordinary activities related to COVID) and epilepsy patients (who neglected visits and sometimes even hospitalization, in order to avoid infection); second, the study protocol, which required two independent physicians with a different level of blinding at each study site and a fixed dose schedule, was perceived as too complex or rigid to apply by a number of potential investigators. As a consequence, of the 28 study sites originally planned, only 18 were activated and 12 actually recruited patients in the trial.
The estimated relapse proportions at 12 months were 47% for the low dose and 48% for the standard dose, with a difference of 1% (95% CI −30%; 27%). No difference in the number or severity of adverse events or quality of life measures was observed between the two treatment groups. Although the small sample size did not allow detection of any significant difference, it is, however, reassuring that the number and severity of adverse events were lower in the low-dose arm.
In the previous research conducted by the PI of the study,5 participants treated with a standard daily dose had only a slightly higher chance of long-term remission than participants receiving low daily doses, but treatment changes for seizure recurrence or adverse events were higher in subjects with standard daily doses rather than in subjects treated with low doses. Drug selection was left to the caring physician's judgment. 122 patients were randomized by 17 centers and followed up to 36 months. 80 cases had complete and evaluable data. 73% of cases receiving low daily doses and 79% of those receiving standard doses entered 24 months remission. Compared to low doses, patients on standard doses had a higher chance of treatment change, mostly due to adverse events.
Performing a literature review, there are no other studies dealing with the standard versus low-dose randomization in patients with a new diagnosis of epilepsy.
In the previous clinical trials about ASMs, no information was provided on the optimal dose or the range of effective dosages.9 The refractory nature of the population and the need to maximize the probability of demonstrating a difference versus controls were two conditions that led to investigating doses in the very high range. For example, 1000 mg/day topiramate, 2400 mg/day oxcarbazepine, and 300–500 mg/day lamotrigine, which were investigated in conversion-to-monotherapy trials of these ASMs, are doses far greater than needed in most patients with previously untreated epilepsy.9, 10
In the review of Perucca et al. 2001,10 the initial target maintenance dosage, defined as the maintenance dosage on which a patient is stabilized after the initial titration phase, is determined by the dose–response profile of the drug and by individual patient characteristics. This dosage is rarely discussed in the literature, and there is no clear consensus because of the lack of specifically designed studies.11 The general approach is for the use of the lowest effective daily dose that is expected to provide seizure control in individual patients, reducing the patient exposure to dosages that are higher than necessary and consequently to higher risks for adverse effects.10
Our findings suggest an equivalence of the low dose versus the standard dose of first-line ASM in the study sample. However, the sample size was too small, and all point estimates obtained were accompanied by wide confidence intervals, crossing the predefined non-inferiority margin. For this reason, no definite conclusion could be drawn, either for the non-inferiority of the low dose or its inferiority or superiority. The research question of the STANDLOW trial remains open, and the difficulty of carrying out a clinical trial in Italy aimed at evaluating the effect of low doses in patients with newly diagnosed epilepsy is confirmed, as it was in the late 90s, made even more difficult by the COVID-19 pandemic.
Possible strategies that could be applied in a future clinical trial to overcome some of the barriers found in the present study includerefining the eligibility criteria, focusing on patients with focal epilepsy of unknown etiology and low seizure frequency; reducing the burden for the study sites by providing a centralized outcome assessment; increasing the number of centres, possibly involving other European countries; and extending the enrolment period to reach the planned sample size.
From a clinical point of view, it should be underlined that the apparent absence of difference was observed in a population of adult patients with newly diagnosed focal epilepsy of unknown or structural etiology, with a low frequency of seizures before diagnosis and an epileptiform EEG in nearly half of the participants. This identifies a specific and clinically relevant population.
ACKNOWLEDGMENTS
This work is dedicated to the memory of Ettore Beghi, who was the PI of this study, and has been our mentor, advisor, and leader over many years and whose participation in this study has been sorely missed. Open access funding provided by BIBLIOSAN.
FUNDING INFORMATION
This research project was financed by the Italian ministry of Health. Project code RF-2016-02363902.
CONFLICT OF INTEREST STATEMENT
E. Bianchi, G. Giussani, S. Beretta, J.C. Di Francesco, M. Cecconi, R. Papetti, E. Rosati, G. Giovannelli, D. Benincasa, M. Longoni, Y. Bartolini, A. Alicino, E. Pronello, V. Cianci, P. Tabaee Damavandi, S. Filipponi, A. Gaiani, S. Diamanti, F. Pasini, G. Pederzoli, T. Francavilla, A. Stabile report no disclosures. M. Maschio has received compensation from EISAI and Ethypharm for participation in conferences. S. Gasparini has received speaker honoraria from Eisai and Biogen. Filippo Dainese has received speaker or consultancy fees from Angelini Pharma, Eisai, and UCB. Pharma outside the submitted work. G. Strigaro received speaker and consulting fees from Eisai and Angelini.
APPENDIX A
Collaborators (STANDLOW study group): Andrea Stabile, Susanna Diamanti, Edoardo Pronello, Giulio Pastorelli, Emilio Davide Arippol, Giovanna Straface, Maria Carlini, Maria Grazia Celani, Teresa Francavilla, Angela Alicino, Vittoria Cianci, Maria Sessa, Stefano Martini.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
