The findings in the above reference contradict findings reported for this agent in many studies dating back to 1990.  Figure 2 in that study shows CD4 decline for the placebo and hydroxychloroquine (HQ) groups. At the viral loads that were treated, the HQ group more closely approximates natural progression of the untreated patient at a CD4 decline rate of 50-75 cells/mm3/year.(1,2) The viral load was slightly higher in the HQ group, but CD4 decline in the placebo group was less than reports of expected rates of decline in untreated individuals. My perception is that the group labeled placebo benefitted from therapy while the group labeled HQ followed the natural course of the untreated.    The ranges in that graph also show that some in the placebo group had an increase in CD4 levels.  CD4 levels at 4 weeks are shown in Fig 2 while the text states that CD4 levels and viral load were not tested at the 4 week visit.

REF
1. Differences in HIV Natural Historyamong African and Non-African Seroconverters in Europe and Seroconverteres in Sub-Saharan Africa. PLosOne.org March 2012 Vol(7) Issue 3. e32369

2.  Predictive Value of Plasma HIV RNA Level on Rate of CD4 T-cell decline in untreated HIV infection.  JAMA Sept 27, 2006, Vol 296 (12) 1498-1506