The null hypothesis of no difference in CR% at 3 months between the arms will be tested against the alternative of a difference in CR% at an alpha level of .05 by assessing the odds ratio for arm yielded by this model.

This is a superiority trial aiming to increase the 3 month complete response rate. The sample size is calculated on the hypothesis that the experimental treatment will increase the 3 months response rate up to 21% (by 3 folds, based on the 7% reported in Scheinberg et al [17]). Under these assumptions, the sample size to reject the null hypothesis is n=96 patients for each treatment arm, increased by 4% for possibly not evaluable patients (total number of 200 patients, 100 each treatment arm). Statistical design for sample size calculation: increase from 7% (control arm) to 21% (investigational arm) in 3 month complete response rate (two-sided binomial test); alpha-error 0.05; power 0.8.

Inclusion Criteria:

1. Diagnosis of severe or very severe aplastic anemia, defined by [29]:

   • At least two of the following:

     • Absolute neutrophil counts <0.5 x 109/L (severe) or <0.2 x 109/L (very severe)
     • Platelet counts <20 x 109/L
     • Reticulocyte counts <60 x 109/L

   • Hypocellular bone marrow (<30% cellularity), without evidences of fibrosis or malignant cells

2. Male or female age > 14 years;

3. Written informed consent

4. Willing and able to comply with all of the requirements and visits in the protocol

5. Understands that they can be randomised to either treatment arm

6. Negative pregnancy test for women of child bearing age

7. Written acceptance to use contraception (hormonal or barrier method of birth control; abstinence) for the entire duration of study participation.

Exclusion Criteria:

1. Prior immunosuppressive therapy with ATG (horse of rabbit) or any other lymphocyte depleting agent (i.e., alemtuzumab)

2. Eligibility to a sibling allogeneic stem cell transplantation

3. Evidence of a myelodysplastic syndrome, defined by the presence of myelodysplastic features, excess of blasts or karyotypic abnormalities typical of MDS (according to revised WHO 2008 criteria) [30],, as well as other primitive marrow disease. Patients with diagnosis of AA with cytogenetic abnormalities which are recurrent in MDS (according to revised WHO 2008 criteria) [30] should be included in this category, and are not eligible for the study; patients with del(20q), +8 and -Y are not included in this category, and thus are eligible for this study. The list of karyotypic abnormalities which qualifies for the diagnosis of MDS are listed in the Appendix.

4. History or clinical suspect of constitutional aplastic anemia (i.e. Fanconi Anemia with positive DEB/MMC test or Dyskeratosis Congenita)

5. History of malignant tumors with active disease within 5 years from enrollment, and/or previous chemo-radiotherapy

6. Previous history of stem cell transplantation

7. Treatment with cyclosporin A unless

   • <4 weeks of cyclosporin A treatment before enrolement and
   • wash out period of 2 weeks before enrollment

8. CMV viremia, as defined by positive PCR or pp65 test

9. WHO performance status ≥3

10. Pregnant or breast feeding patients

11. Patients with hepatic, renal or cardiac failure, or any other life- threatening concurrent disease

12. Patients with HIV infection

13. Patients without social health care assistance

14. Participation in another clinical trial within 1 month before the start of this trial

15. Patients and/or female partners of male patients not using highly effective method of birth control i.e. intrauterine device (IUD), hormonal (oral pill, injection, implants), tubal ligation or partner's vasectomy

16. subjects with known hypersensitivity to any of the component medications

The presence of a Paroxysmal Nocturnal Hemoglobinuria clone is not an exclusion criterion.