Neurofibromatosis Type 2
The NFTI-QOL's eight-question snapshot plus the hearing, tinnitus, and balance of every day in between.
NF2 tumors erode hearing, balance and facial function slowly and unpredictably, yet the outcomes that matter most are measured only at an audiogram and MRI every 6 to 12 months. Between those two dots, patients live the real curve. Forma captures it in their own voice, day by day.
Forecast
Based on prior Topics logging
Topics
How it works
15 seconds of speech. 25 data points.
Forma turns each patient into a continuous data stream. In under a minute, by voice or text, patients log hearing change per ear, tinnitus, balance and falls, facial-nerve and vision changes, tumor-related pain, mood and medication, and rate the eight NFTI-QOL domains as a repeated, time-stamped signal instead of a once-a-visit snapshot. The result is a sensitive, multi-domain PRO stream that pairs self-reported hearing with objective audiometry and directly attacks the sensitivity-to-change problem limiting NF2 trial endpoints. It plugs into existing surveillance rather than replacing it.
“”
Forecast
Based on prior Topics logging
Topics
Structured, coded, longitudinal
- Timing
- Seizure time:6:00 AMSeizure duration:45 secRecovery time:20 minMissed dose time:Previous evening
- Location
- Injury location:Tongue
- Severity
- Injury:Tongue bitePost-ictal symptom:Confusion
- Actions
- Adherence:Missed doseMedication:Keppra
- Environmental
- Barometric pressure:29.82 inHg (falling)Moon phase:Waxing gibbous (87%)
From data to insights
What the data reveals
Daily voice logs compound into actionable insights, for patients managing their condition and for researchers running studies.
For Patients
Is my right ear actually getting worse, or does it just feel that way?
Across the last 6 weeks you rated right-ear hearing worse on 9 days versus 2 for the left, and 'phone calls' was your most-logged difficulty (14 times). The trend was downward until your first infusion on Jun 12, after which it flattened.
Right-ear difficulty
Did the bevacizumab actually help my tinnitus?
Your tinnitus loudness averaged 8/10 in the two weeks before treatment and 5/10 in the two weeks after, and sleep-disruption logs fell from 6 nights to 2.
Tinnitus loudness
When is my balance worst so I can plan around it?
78% of your dizziness logs were triggered by head movement or low light, and every near-fall (3 total) happened after dark. Mornings and well-lit rooms were almost always steady.
Falls risk by light level
Can you summarize my last few weeks for my doctor?
Between Jun 8 and Jul 1, Marcus started bevacizumab (first infusion Jun 12) for a growing right vestibular schwannoma. Self-reported right-ear hearing improved from a moderate to a mild problem and the Jun 29 audiogram confirmed a stable-to-improved word score; tinnitus loudness fell from ~8 to ~5 with sleep disruption down from 6 to 2 nights; mid-back tumor pain eased from 6/10 to 2/10; blood pressure required a new antihypertensive; mood recovered from severe scanxiety (Jun 10) to mild by Jul 1.
For Researchers
Does Forma's daily hearing PRO detect decline earlier than the surveillance audiogram?
In a pilot cohort (N=58), self-reported per-ear hearing worsening preceded a documented word-recognition drop by a median of 23 days (IQR 12–41), with sensitivity 0.81 and specificity 0.74 against the next audiogram.
Lead time (days)
Trend
Is the repeated NFTI-QOL signal more sensitive to change than a single visit administration?
Repeated in-app NFTI-QOL domain scores showed a standardized response mean of 0.71 over 12 weeks versus 0.34 for paired clinic administrations (N=44), reducing the ceiling effect that limits single-visit administration.
Responsiveness (SRM)
Trend
Does patient-reported hearing correlate with objective word recognition?
The Word Understanding proxy correlated with clinic word-recognition scores at r=0.68 (p<0.001, N=52), supporting its use as a between-visit surrogate paired with audiometry.
PRO vs audiometry
Trend
Can the platform flag treatment response for bevacizumab candidates?
Among bevacizumab starters (N=27), a composite of tinnitus loudness and hearing-change PROs separated responders from non-responders by week 4 with AUC 0.79, weeks ahead of the confirmatory audiogram.
Response separation
Trend
The raw signal
11 days of raw patient data
The day-by-day voice log behind the insights above, and the detail that periodic check-ins can't capture.
Baseline before treatment: right-ear hearing declining and hard on the phone, tinnitus loud and disrupting sleep, feeling anxious ahead of his scan.
Surveillance MRI and skull-base consult; the right-side tumor has grown and the team recommends starting off-label bevacizumab to protect hearing. Scanxiety peaks.
First bevacizumab infusion, about ninety minutes, no infusion reaction.
Post-infusion fatigue and blood pressure up on the new antihypertensive; mood still low but steady.
Tinnitus in the left ear a notch quieter than baseline; sleeping a little better.
Second (loading) infusion; blood pressure up at the chair and monitored, otherwise uneventful.
Right ear feels a bit clearer for the first time in months and the mid-back tumor pain has eased.
Mood lifting, noticeably less anxious now that he can hear on the phone again.
Maintenance infusion with a mild headache afterward; back pain remains improved.
Audiogram confirms right-ear word recognition has held and edged up; hearing the best it has been this quarter.
Overall check-in is the strongest in weeks: tinnitus quieter, pain low, mood recovering.
Key insight
Over three weeks, Marcus's tinnitus dropped from 3 to 2 and his self-reported hearing problem eased from 2 to 1 after the first two bevacizumab infusions (Jun 12 and Jun 19), with the Jun 29 audiogram confirming his right-ear word score had held. Mood spiked to severe with scanxiety on Jun 10 and then recovered to mild by Jul 1 as hearing returned. None of this between-infusion trajectory is visible at a twice-yearly clinic visit.
See Neurofibromatosis Type 2 configured for your study
See Forma configured for NF2