Belinostat, Carboplatin and Paclitaxel (BelCaP) Compared to Carboplatin and Paclitaxel in Patients With Cancer of Unknown Primary

Inclusion Criteria: * Patients with CUP where the primary site had not been revealed by complete history, physical examination (including gynecological examination when appropriate), computed tomography (CT) scan of the chest, abdomen and pelvis, bilateral mammography (in women with adenocarcinoma or poorly differentiated carcinoma), routine laboratory studies (complete blood cell counts, electrolytes, urinalysis, liver and renal function tests), and directed work-up of any other symptomatic areas. * Light microscopic pathologic diagnosis of adenocarcinoma (including poorly differentiated), squamous cell carcinoma, or poorly differentiated carcinoma. Patients with poorly differentiated carcinoma must have immunohistochemical stains to confirm the diagnosis of carcinoma, and to rule out other tumor types. Note: patients with a light microscopic histology diagnosis of "poorly differentiated neoplasm, not otherwise classified" did not fulfill the criteria for inclusion, unless immunohistochemical staining confirmed the diagnosis of carcinoma. * Signed consent of an IRB (\[Institutional Review Board\])/IEC (\[Independent ethics committee\]) approved ICF (\[Informed Consent Form\]). * At least one measurable lesion according to RECIST (\[response evaluation criteria in solid tumors \]) criteria. Note, target lesions could only be selected within previously irradiated areas if newly arising or clearly progressing after irradiation as proven by repeat scanning * Performance status Eastern Cooperative Oncology Group (ECOG) ≤ 2. * Age ≥ 18 years. * A negative serum or urine pregnancy test for women of childbearing potential. Postmenopausal women must have been amenorrheic for ≥ 12 months to be considered of non-childbearing potential. * Serum potassium within normal range. * Acceptable coagulation status: Prothrombin time/International normalized ratio PT/INR (\[international normalized ratio\]), and activated partial thromboplastin time (APTT) ≤ 1.5 × upper limit of normal (ULN) or in the therapeutic range if on anticoagulation therapy. * Acceptable liver, renal and bone marrow function including the following: 1. Bilirubin ≤ 1.5 times ULN (if liver metastases were present, then ≤ 3 × ULN was allowed). 2. Aspartate aminotransferase/serum glutamic oxaloacetic transaminase (AST/SGOT), alanine amino transferase/serum glutamic pyruvic transaminase (ALT/SGPT), and alkaline phosphatase ≤ 3 times ULN (if liver metastases were present, then ≤ 5 × ULN was allowed). 3. An estimated creatinine clearance ≥ 45 mL/min using an appropriate formula (Appendix C, protocol version 1.0, Appendix 16.1.1), or measured ethylenediaminetetraacetic acid (EDTA) renal clearance ≥ 45 mL/min. 4. Absolute neutrophils count ≥ 1.5 × 109/L, platelets ≥ 100 × 109/L. 5. Hemoglobin ≥ 9.0 g/dL or ≥ 5.6 mmol/L (patients with chronic anemia due to underlying disease and its treatment could undergo blood transfusion prior to treatment in order to meet this criteria). Exclusion Criteria: * Patients with well recognized subsets of CUP site where treatments directed towards a defined tumor type, or surgery, alternatively radiotherapy, can be advised: * Women with adenocarcinoma involving only axillary lymph nodes. * Women with papillary serous carcinoma of the peritoneum. * Women with adenocarcinoma with positive staining for estrogen receptor (ER) or progesterone receptor. * Young men (\< 45 years) with poorly differentiated carcinoma consistent with an extragonadal germ cell tumor (carcinoma involving mediastinum or retroperitoneum, or elevated levels of beta-human chorionic gonadotropin or alpha-fetoprotein). * Men with bone metastases and/or adenocarcinoma, and abnormally elevated PSA (\[Prostate specific antigen\]) in their plasma. * Patients with squamous cell carcinoma involving only cervical lymph nodes, or inguinal lymph nodes. * Patients with neuroendocrine carcinomas determined according to standard pathology diagnosis procedures, including stains. * Patients with potentially completely resectable metastatic disease, or disease which can be adequately treated with radiotherapy only. * Patients with brain or meningeal metastases. Note, patients with adequately treated brain metastases, e.g. surgically resected, or adequately controlled by radiotherapy, with no residual neurological symptoms due to metastases and no steroid treatment required, could be enrolled. If clinical suspicion, adequate investigations should be performed to rule out brain metastases or meningeal involvement. * Prior systemic anti-tumor therapy, including chemotherapy administered in association to radiotherapy for sensitization, for CUP. Note, prior radiotherapy or surgery was allowed provided treatment was completed at least 4 weeks before randomization. * Treatment with investigational agents, including non-anti-tumor agents, within the last 4 weeks before randomization. * Co-existing active severe infection or any co-existing medical condition assessed by the Investigator as likely to interfere with study procedures. * Significant cardiovascular disease (New York Heart Association Class III or IV cardiac disease), myocardial infarction within the past 6 months, unstable angina, unstable arrhythmia or a need for anti-arrhythmic therapy (use of medication to control heart rate in patients with atrial fibrillation was allowed, if on stable medication for at least the last month prior to randomization and the medication not listed as causing Torsade de Points (Section 13.2, Appendix B, protocol version 1.0, Appendix 16.1.1), or evidence of acute ischemia on ECG. * Marked baseline prolongation of QT/QTc (\[corrected QT interval\]) interval, i.e., demonstration of a QTc interval \> 450 millisecond (ms); Long QT Syndrome; the required use of concomitant medication that may cause Torsade de Pointes (Section 13.2, Appendix B, protocol version 1.0, Appendix 16.1.1). * Altered mental status precluding understanding of the informed consent process and/or completion of the necessary study procedures. * History of a previous malignancy within 5 years with the exception of non-metastatic non-melanoma skin cancer or cervical carcinoma in situ. Prior systemic therapy for other malignancy completed at least 5 years before randomization is allowed. Implemented with amendment 2 (study centers in France only): History of a previous malignancy, irrespective of time since diagnosis/treatment, with the exception of non metastatic non-melanoma skin cancer or cervical carcinoma in situ. * Known hypersensitivity to either platinum compounds or paclitaxel, or any components of the study medications, and inability for desensitization. * Known infection with HIV, or known active Hepatitis B or C infection. * Peripheral neuropathy ≥ Grade 2. * Pregnant or lactating females. * Women of childbearing age and potential who are not willing to use effective contraception during the study and until 30 days after last dose of study drug. Male patients or male patients who have female partners of childbearing age and potential who are not willing to use effective contraception during the study and until 30 days after last dose of study drug. Highly effective methods of birth control are defined as those which result in a low failure rate (i.e. less than 1% per year) when used consistently and correctly such as implants, injectables, combined oral contraceptives, some intra uterine devices, sexual abstinence or vasectomized partner. * Patients that are not affiliated with social security (study centers in France only). * Implemented with amendment 1 (study centers in Denmark only): Hearing impairment assessed by the Investigator as being of such a degree that treatment with carboplatin cannot be initiated. * Implemented with amendment 1 (study centers in Denmark only): Bleeding tumors.