Memantine (Namenda) Autism Protocol

​What We Measure and Why (Memantine Protocol)

Frontier Health and Wellness (FHW) employs a measurement-based care approach in its memantine treatment protocol for autism. This ensures that clinical decisions are driven by objective changes over time rather than guesswork. The tools described below are not “tests” in the evaluative sense, nor are they used to judge a child’s abilities. Instead, each instrument is used to establish a baseline and track specific symptom domains at regular intervals, allowing us to compare pre-treatment patterns to follow-up data. By collecting quantifiable metrics before treatment and at subsequent check-ins, we are better able to discern true treatment effects and adjust care accordingly, rather than relying on subjective impressions alone.

​

SRS-2 (Social Responsiveness Scale)
The Social Responsiveness Scale, Second Edition (SRS-2) is a validated caregiver rating scale that quantifies social communication abilities and social interaction patterns in children with ASD. We include the SRS-2 in our protocol to objectively measure changes in social engagement, reciprocal interaction, and social withdrawal – key domains that memantine is intended to improve. This parent-completed questionnaire yields a standardized score reflecting the child’s social responsiveness. The rationale for its inclusion is that meaningful improvements in day-to-day social functioning (e.g. more back-and-forth communication or reduced social avoidance) would imply the treatment is having the desired effect. The aim is not to achieve a “perfect” or neurotypical score, but rather to observe clinically significant shifts from the child’s own baseline functioning in the targeted direction.

In practice, SRS-2 scores guide decision-making by providing a quantitative gauge of progress. For example, we look for a substantial reduction in the SRS-2 total score from baseline as evidence of improvement. In our memantine protocol, a ≥25% improvement on the SRS-2 (total score) is one criterion for continuing the medication into the maintenance phase. Conversely, if after an adequate trial period (e.g. >8 weeks at the target dose) there is minimal or no change in the SRS-2 score, it may suggest a lack of efficacy – supporting a decision to reassess or discontinue the treatment. By tracking the SRS-2 at baseline and regular follow-ups, we ensure that any changes in social responsiveness are captured objectively.

​

CGI-I (Clinical Global Impression – Improvement)
The Clinical Global Impression – Improvement (CGI-I) scale is an industry standard clinician- rated measure that provides a global assessment of the patient’s overall improvement or worsening relative to the treatment start. Essentially, at each evaluation the clinician asks: “Compared to where we started, is the child meaningfully improved, unchanged, or worse?”.
The CGI-I condenses input from multiple sources – parent reports, direct clinical observations, and the child’s functioning across settings – into one consistent clinical anchor of change. This tool is widely used in psychiatric and behavioral research and health practice, especially in medication trials, because it offers a standardized way to capture the overall magnitude of change in a patient’s condition.
Including the CGI-I ensures that a professional’s objective impression of change is formally recorded alongside caregiver-rated scales. We establish a baseline anchor (CGI-I is defined as “no change” at the outset) and then assign a CGI-I score at each follow-up to rate how the child’s overall symptoms compare to baseline. The CGI-I is helpful for decision-making: a favorable CGI-I rating (for instance, “much improved” or better) provides support that the treatment is benefitting the child, while a rating of “no change” or worse raises concern that the treatment may not be effective. In our memantine protocol, if a child is rated as at least minimally improved or better (CGI-I score ≤ 3) at key checkpoints, this is considered evidence of a positive clinical response warranting continuation of memantine. On the other hand, if the CGI-I indicates no appreciable improvement after a full titration period, we would carefully reconsider the plan (potentially tapering and discontinuing memantine). By using the CGI-I, we incorporate a broad clinical perspective that complements the specialized symptom measures.

​

GPS-A (Glutamatergic Profile Screener for Autism)
The Glutamatergic Profile Screener for Autism (GPS-A) is a research-informed screening and monitoring tool developed at FHW to identify a specific neurobehavioral profile semi-characterized in research to be more responsive to memantine. This parent-rated checklist captures the symptom pattern associated with an excitatory-inhibitory imbalance in ASD –sometimes referred to as a “glutamatergic subtype.” Children with this phenotype often exhibit:

• Social and Emotional Modulation difficulties (capable of social engagement when calm, but prone to withdraw or have intense emotions when overwhelmed).

• Sensory Over-reactivity (e.g. extreme sensitivity to sounds, lights, touch)

• Elevated Excitatory Arousal and slow recovery from stimulation (a physiologic “over‐switched-on” state with persistent tension or arousal even when calm)

• Cognitive and Behavioral Reactivity to overload (racing thoughts or repetitive looping under stress, irritability or meltdowns when overstimulated, followed by exhaustion)

• Medication and Stimulant Sensitivity (atypically strong or agitated responses to stimulants, SSRIs, caffeine, etc.)

​

The GPS-A checklist assigns numeric ratings to specific domains and yields a total score (0–60) indicating how prominently the glutamatergic/excitatory phenotype is expressed. The higher the score, the more the child’s profile aligns with the target subtype. We emphasize that the GPS-A is a research-informed instrument currently under internal clinical evaluation – it is not yet a formally validated diagnostic tool and cannot “prove” that memantine will work for a given child. Rather, we use it to enhance our clinical reasoning: a high GPS-A score suggests the child fits a higher-likelihood responder profile, supporting the decision to proceed with a structured memantine trial, whereas a low score might prompt consideration of alternative interventions.

Rationale: The inclusion of GPS-A is grounded in emerging research on glutamate’s role in an ASD subpopulation. Evidence from clinical studies indicates that children with signs of cortical hyper-excitability linked to NMDA-receptor dysfunction show notably higher response rates to memantine than those without this profile. For instance, one analysis found an ~80% response rate to memantine in children identified with this excitatory glutamate pattern, compared to ~20% in unselected patients. Since direct biomarkers (like specialized brain imaging or MR spectroscopy of glutamate levels) are not readily available in routine practice, the GPS-A serves as a practical proxy to flag this phenotype. In other words, it operationalizes real-world
observations into a quantifiable score that we can track.
Use in Monitoring: Beyond screening eligibility, we also track the GPS-A score over the course of treatment to see if the targeted symptom cluster is shifting. A baseline GPS-A total score ≥30 (out of 60) is one of our eligibility criteria, indicating a “probable or possible glutamatergic subtype” suitable for memantine treatment. During the trial, if we observe a reduction in the GPS-A total score or if the parent’s qualitative descriptions reflect improvement in these excitatory-inhibitory symptoms, it implies that memantine is impacting the intended neural systems. Such improvement – for example, better sensory tolerance, easier calming, or improved social responsiveness – would reinforce our decision to continue treatment. In contrast, if the GPS-A scores remain high and the child’s excitatory symptoms persist unchanged, that lack of change would weigh into a decision to adjust the approach or discontinue memantine for that patient. The GPS-A data, therefore, provide an additional layer of phenotype-specific monitoring that complements the broader SRS-2 and CGI-I measures.

​

Why Consistent Measurement Matters

Autism symptoms can naturally fluctuate from week to week due to factors like sleep variations, stress levels, school or routine changes, illness, or other life events. Without consistent measurement, it is easy to misinterpret these normal fluctuations – or a placebo effect – as evidence that a treatment is working or not working. By using the same tools in a structured, repeatable manner, we improve the signal-to-noise ratio in our assessments. In short, consistent measurement helps us separate true treatment effects from background variability. This rigor is especially important in medication trials where expectation bias (on the part of family or clinician) can otherwise cloud judgment. Maintaining a measurement schedule with validated instruments creates an objective record of the child’s trajectory.

Some key reasons we use this measurement-based approach include:

• Minimizing placebo attribution: By relying on objective rating scales, we reduce the chance of crediting memantine for changes that are actually due to expectation or coincidence. Consistent data collection helps ensure that perceived improvements are backed by measurable change rather than placebo effect.

• Reducing false signals from environment or development: Standardized measures taken at regular intervals help distinguish true drug-related improvements patterns from symptomatic ups and downs caused by external stressors or developmental ebb and flow. For example, if a child’s behavior improves because school is out for the summer (reduced stress) or worsens temporarily due to a cold, the larger pattern established through the assistance of our tracking tools help us recognize those context effects so we don’t misattribute them to the medication.

• Standardizing side effect evaluation: We use consistent checklists and recommend caregiver logs to monitor side effects (sleep, appetite, mood changes, etc.) at each visit. This uniform approach means any emerging adverse effect is noted and interpreted in context, rather than being overlooked or confused with symptoms of the condition. It improves safety by ensuring we catch patterns (e.g. increased irritability or insomnia) early and attribute them correctly – either to memantine or to unrelated factors.

• Informing clear data-driven decisions: Perhaps most importantly, ongoing measurement provides concrete criteria for when to continue, adjust, or stop the trial. We set predefined benchmarks (like the SRS-2 and CGI-I thresholds mentioned above) 
  that assist in determining treatment effect. If those benchmarks aren’t met, the data give us confidence in pivoting the treatment plan. Likewise, if clear improvements are documented, we have justification for pursuing the therapy further. This reduces 
  ambiguity for both the clinician and the family, as decisions are backed by evidence from the tracked outcomes.

​

FHW’s memantine protocol is designed with rigorous outcome tracking at its core. The approach aligns with best practices in measurement-based psychiatry, enhancing the reliability of our clinical assessments. By combining caregiver-reported measures (SRS-2, GPS-A) with clinician-rated impressions (CGI-I), and by monitoring these consistently over time, we strive to provide a high standard of care that is both personalized (phenotype-informed) and evidence-guided. The ultimate goal is to ensure that if memantine is helping a child, we recognize that improvement with an enhanced confidence – and if it’s not, we have the clarity to redirect treatment in a timely manner, all while maintaining safety and objectivity in the therapeutic process.