Placebo for AFQ056 will be supplied as oral tablets.

Patients randomized will be up titrated to the target doses at regular intervals during the up titration period.

AEs will be summarized by presenting, for each treatment group the number and percentage of patients having any adverse event, having an adverse event in each system organ class and having each individual adverse event by system organ class and preferred term.

The LFADLDS assesses the degree of dyskinesia interfering with activities of daily living. Specific definitions for severity rating codes (range, 0-4 for each task) will be provided for reproducibility of results. A higher score indicates more severe impairment. A patient and caregiver version of the revised LFADLDS will be used in this study.

Change from baseline to Week 12 will be analyzed using MMRM including treatment, pooled center, week, treatment by week interaction as fixed effects and baseline total score as covariate with an unstructured covariance structure.

The CGIC provides investigator-rated assessment of change from baseline to assess disability due to dyskinesia. Change from baseline will be rated on a 7-point, Likert-type scale where 1 = markedly improved, 2 = moderately improved, 3 = minimally improved, 4 = unchanged, 5 = minimally worse, 6 = moderately worse, and 7 = markedly worse.

The CGIC score at week 12 last observation carried forward (LOCF) will be analyzed using analysis of covariance (ANCOVA) model with treatment group, pooled center, and baseline CGIS (Clinical Global Impression of Severity) score as factors.

A patient home diary developed and validated for use in PD patients will be used to record whether the patient is asleep, OFF, ON without dyskinesia, ON with troublesome dyskinesias, and ON with non-troublesome dyskinesia.

Change from baseline to Week 12 will be analyzed using MMRM including treatment, pooled center, week, treatment by week interaction as fixed effects and baseline total score as covariate with an unstructured covariance structure.

UPDRS assesses the disease state of PD. Question 32 assesses length of dyskinesias in percentage of the day. Question 33 assesses disability during the previous week (not disabling, mildly disabling, moderately disabling, severely disabling, completely disabling). Question 34 assesses painfulness of dyskinesias from 0 (no painful dyskinesias) to 4 (marked).

The change will be analyzed using MMRM including treatment, pooled center, week, treatment by week interaction as fixed effects and baseline total score as covariate with an unstructured covariance structure.

Pulse and blood pressures are taken at each visit. Vital sign data will be summarized by presenting summary statistics for change from baseline values. The incidence rates of clinically notable vital sign abnormalities will be summarized.

Standard hematology with differential; measures of coagulability: aPTT, PT/INR; standard clinical chemistry; FSH, LH, oxytocin, prolactin, TBG, TSH, and T4; urinalysis (specific gravity, protein, glucose and blood) Laboratory data will be summarized by presenting shift tables, by presenting summary statistics, change from baseline values, and incidence rates of clinically notable abnormalities summarized.

A standard 12-lead ECG will be performed. A central facility will be used for interpretation and analysis of the ECGs.

ECG intervals will be summarized by summary statistics for change from baseline values and incidence rates of clinically notable abnormalities summarized.

The occurrence of adverse events would be sought by non-directive questioning of the patient. Adverse events may be detected when they are volunteered by the patient or through physical examination, laboratory test, or other assessments.

Treatment-emergent adverse events (TEAEs) will be summarized by presenting, for each treatment group, the number and percentage of patients having an AE, having an AE in each system organ class and having each AE by system organ class and preferred term.

Part III of the UPDRS (items 18-31; score 0-56), measures speech, facial expression, tremor, action or postural tremor, rigidity, finger taps, hand movement, alternating movement, leg agility, arising from a chair, posture, gait, postural stability, and bradykinesia. A higher score means worsening of symptoms.

Changes from baseline to Week 12 will be analyzed using MMRM including treatment, pooled center, week, treatment by week interaction as fixed effects and baseline total score as covariate with an unstructured covariance structure.

The MMSE is a test of cognitive dysfunction consisting of orientation, registration, attention-calculation, recall, and language administered by a health care professional. The MMSE results in total possible score of 30, with higher scores meaning better function.

The change from baseline to endpoint at Week 12 in MMSE total score will be analyzed using LOCF ANCOVA with treatment group and pooled center as factors and the baseline score as a covariate. Treatment differences in the change from baseline for scores with 90% CI will be reported as descriptive statistics.

This test consists of: Detection Task, Identification Task, One Card Learning Task, One Back Task, and International Shopping List Task. Speed times and working memory will be measured by the number of correct responses.

The change from baseline to endpoint at Week 12 for cognitive function will be analyzed using LOCF ANCOVA with treatment group and pooled center as factors and the baseline score as a covariate. Treatment differences in the change from baseline for cognitive function with 90% CI will be reported as descriptive statistics

The SCOPA-PC is an easily administered semi-structured, questionnaire developed for the assessment of psychiatric symptoms in Parkinson's disease patients administered by a clinician with input provided by patient and caregiver. The total SCOPA score ranges from 0-21, with higher scores reflecting more psychiatric complications.

The SCOPA-PC total score and each of the item scores will be summarized.

The C-SSRS assesses suicidal ideation/behavior using a patient interview. The data is mapped to Columbia Classification Algorithm for Suicide assessment. The code and categories are: completed suicide, suicide attempt, preparatory actions toward imminent suicide behavior, suicidal ideation, self-injurious behavior without suicidal intent.

The proportion of patients who are coded in the categories above, the proportion of patients with any suicidal behavior engaged in during the study, and the proportion of patients with suicidality will be summarized.

Blood samples will be taken by either direct venipuncture or an indwelling cannula inserted in a forearm vein. The first blood sample will be collected 1-3 hours after morning study medication dose. The second sample will be collected 3-5 hours after the morning study medication dose. The third blood sample will be collected 5-7 hours after the morning study medication dose.

Summary statistics for concentrations per sampling window will be provided and average plasma levels will be calculated.

Blood samples will be taken by either direct venipuncture or an indwelling cannula inserted in a forearm vein. An alternative sampling method will also be explored. At all visits, the sample will be collected 1-3 hours after morning study medication dose. At Week 12, the second sample will be collected 3-5 hours after the morning study medication dose and the third sample will be collected 5-7 hours after the morning study medication dose.

Summary statistics for concentrations per sampling window will be provided and average plasma levels will be calculated.

Investigation of the safety of AFQ056 with concomitant amantadine is being studied to gather drug-drug interaction data of amantadine and AFQ056 in this patient population.