ISCTM 7th Annual Scientific Meeting – Negative Symptoms Working Group Dinner – Summary
Negative Symptoms
Working Group Dinner
21 Feb 2011
Chairs: S Marder, D Daniel
1. To establish persistent and predominant negative symptoms a 4-6 week prospective lead-in is recommended, along with a retrospective look back period. The latter is difficult to document in a multicenter industry sponsored clinical trial setting.
2. Requiring a minimum level of negative symptoms for inclusion could invite baseline inflation by investigators , but is necessary to establish an inclusion criteria of persistent predominant negative symptoms.
3. To establish a minimum inclusion score of negative symptoms for entry a CGI-negative symptoms score of moderate or greater and/ or an average score of moderate or greater on the PANSS negative factor would be acceptable. .
4. The group did not establish a consensus on whether a maximum limit of positive symptoms should be an inclusion criteria for negative symptom trials. The group agreed that requiring “predominant” negative symptoms may be too limiting. “Prominent” may be better. A CGI-positive symptom score of mild or less and/or a PANSS positive subscale average score of mild or less were mentioned as measures that could be useful in this regards. Specific positive symptom item scores could also be utilized.
5. Baseline inflation by investigators may be addressed by : 1) using a different scale for inclusion criteria vs. primary outcome; 2) variable length placebo lead- in prior to randomization; 3) surveillance of ratings and regular feedback to raters; (4) use of remote raters who won’t be biased toward inclusion
6. What is the threshold for “clinically meaningful” change? For registration trials a global measure should be combined with an effect size of at least .5 on either the PANSS Marder factor, NSA-16 or SANS. Among global measures a CGI keyed to negative symptoms is among the acceptable global measures for this purpose. This combination would also be useful in proof of concept studies, although more flexibility is usually appropriate in a proof of concept study..
7. A relatively homogenous patient sample is acceptable in proof of concept studies. In contrast,in registration studies a less homogenous sample that is more likely to generalize is more appropriate.
8. In selecting a measure of functional improvement in negative symptom studies it is important to avoid a functional scale that reproduces or largely overlaps with the negative symptom scale. The QLS strongly correlates with negative symptom scales but is measuring largely the same thing. The SFS is also redundant to negative symptom scales. The SOFAS and PSP are functional scales that are relatively less redundant to negative symptom scales.
9. The CAINS and BNSS are the products of separate but overlapping work groups that grew out of the NIMH-MATRICS Consensus Development Conference on Negative Symptoms. They incorporate all 5 currently recognized domains of negative symptoms. They include both anticipatory and consummatory anhedonia, and separate behavior and internal experience. They are designed to primarily rely on information obtained by interview of the patient. The BNSS is shorter and is available currently for clinical trial use, probably most appropriately as an exploratory variable. It has been used in relatively few subjects, thus far. The CAINS has been tested in more patients but is undergoing continued revision and field testing and currently has limited availability for clinical trials.
10. Data presented by Roche indicates that negative symptoms are a major contributor to psychopathology in about 55% of patients seen in their long-term trials.
11. Important characteristics of an instrument for measuring negative symptoms include: valid across countries and cultures; sensitive to change; simple and short.