Antidepressant Responses and Segregation Analyses in Affective Families
Angela Orsini
Institute of Clinical Psychiatry, Milan Medical School, Ospedale S. Paolo, 20142 Milan, Italy
Previous segregation analysis studies have pointed to different genetic susceptibility transmission modes underlying similar affective clinical manifestations. In our previous study (13) and in literature reports (5), the application of the Single Major Locus (SML) and the Multifactorial Polygenic (MFP) hypotheses of transmission to the observed frequency of Affective Disorders in first-degree relatives of affective patients has not been very satisfactory. It has not enabled us to rule out either of these hypotheses, perhaps because neither one takes information about second-degree relatives into account. The following application of segregation analyses to the same data (second-degree relatives included) (14) allowed us to test only some genetic hypotheses, those involving modified SML, the dominant and the recessive ones. The analyses were conducted both including (4) and excluding (14) data on secondary cases belonging to the Affective Spectrum Disorders. These studies made it possible to identify some groups that have similar intra- but not interfamilial segregation structures. In the present work we tested different and antithetical transmission mechanisms, also taking other segregation hypotheses of familial susceptibility into consideration. In agreement with the suggestion of the existence, within our sample, of a dishomogeneity in choosing familial transmission mode, we attempted to subdivide family groups on the basis of a clinical criterion, outcome on tricyclic drugs. This is an approach strictly related to the pathogenetic and biological aspects of Affective Disorders and to the familial load, as has been recently reported by Prusoff et al. (11). In this work, a genetic family study of nonbipolar depressed patients, subtypes of depression were identified on the basis of the speed and magnitude of drug responses, and the family data were used to validate these classifications. The first-degree relatives of patients with the most rapid drug responses had the highest rate of depression (23.7%), whereas those subjects with the slowest first week treatment responses showed the lowest rates of depression (10.8%). At treatment termination (after 4 weeks), first-degree relatives of the patients who had had an intermediate response had significantly more depression (23%) than those whose probands were rapid (13%) or slow (11%) 4-week responders. It is well known that, despite the success of antidepressant tricyclics and the prophylactic responses to lithium salts, poor treatment outcome does occur in some patients. At this time, neither simple pharmacokinetic mechanisms nor clinical or semiological patient characteristics have been able to explain these negative outcomes. The present study was designed to test the relationship between the familial susceptibility segregation modes of the disease and the possible genetic mechanism underlying the proband outcome on antidepressant treatments. The presence or absence of a drug response may differentiate biologically and/or biochemically distinct forms of Affective Disorders that may be genetically (and thus ecologically) different. The aim of our study was the identification of the basis for heterogeneity of Affective Disorders by studying the segregation structures in a sample of families representative of the whole affective population. This was our rationale for selecting tricyclic therapy outcome, rather than lithium response, as our indicator, since probands who had been under lithium therapy had shown a high rate of recurrence that is characteristic of a specific Affective Disorder subgroup. In genetic research an intriguing problem is the definition of the affected phe-notype. Previous epidemiological studies (6,10,15) showed that some diagnostic categories, which are not usually part of Major Affective Disorders, such as Dysthymic Disorders, Cyclothymic Disorders, and Atypical Depressions, when occurring in informative families, could be incomplete forms (intermediate phe-notypes) of the same disease. A combined epidemiological study (10) of Swedish and Italian probands with Affective Disorders indicated that, when also taking into account those diagnoses, the morbidity risks in the two populations were of the same entity and were affected by the same factors from a clinical and epidemiological point of view. From the genetic viewpoint we assumed that these disorders belonged to the Affective Spectrum Disorders and therefore were clinical manifestations of the same genotype.