Amendment 2 was prepared to implement the Urgent Safety Measures in terms of study drug administration which included: 1. Dosing schedule modification: the dosing schedule was modified to an every other week (QOW) dosing with a cycle length of 28 days. For patients who were well tolerating reduced doses, dose re-escalation was also allowed in order to seek maximal clinical benefit from panobinostat. 2. Administration of panobinostat with food: the results of another study, [CLBH589B2111], indicate that panobinostat could be administered without regard to food. These findings were consistent with the results from an earlier pilot food effect [CLBH589B2101] study where it was even recommended to take food with panobinostat as it helps decreasing GI toxicities. 3. Modifications of the ECG schedule: based on updated data from other panobinostat clinical trials, protocol amendment 2 reduced the cardiac monitoring of this study and thus reduced the number of ECGs that were needed to be performed during the course of the study. 4. Patient population extension: all participating sites were contacted iregarding the slow enrollment rate. The contacted investigators indicated that they would favor inclusion of arm I patients with up to two prior cytotoxic chemotherapy in the metastatic setting as there was a medical need for this patient population. Amendment 2 extended the patient population and allowed ER+ and/or PgR+ patients with 2 lines of prior cytotoxic chemotherapy in the metastatic setting to be eligible for the trial.

Protocol amendment 3, included the following: 1. Closure of enrollment in Arm II: In Arm II (the ER- and PgR- patient population), accrual was slower than anticipated and the required number of evaluable patients had not been reached in stage I (only one partial response had been observed in the ongoing patients). Because of the very low patient enrollment and because the Risks to Benefit ratio seen thus far did not justify the accrual of more patients in this arm, a decision was made to close this study arm for enrollment. The patients already included were given the possibility to continue in the study. It was worth noting that in Arm I (the ER+ and/or PgR+ patient population), the enrollment was temporarily closed as per protocol as the first cohort was enrolled and the number of responses required to open stage II was not reached (only one response was observed whereas three were required to open stage II). 2. Follow-up: because of the small number of patients registered in the study (premature end of study arm II, and stage II not opened for arm I), the survival information would not have provided information that would have been relevant for the patients and for panobinostat development. For this reason, the follow-up part of the protocol was removed and the patients were not followed after their End-of-Study visit. 3. Secondary and exploratory objectives of the study: with such a small number of patients registered in the study, independent central review of the scans would not have provided statistically relevant information, nor would the assessment of progression-free survival, time to response, duration of response, overall survival, or the confirmation of the patients’ status of HER2, Estrogen Receptors and Progesterone Receptors by a central laboratory. These three objectives and their corresponding endpoints were therefore removed from the protocol. The exploratory objective performed on core needle biopsy was also removed.