Mutations at positions 151, 65 (intermediate/high level resistance), 215, 184, 75, 74, 69 (low level resistance), 219, 210, 118, 116, 77, 70, 67, 62, 44 and 41 (contributes to resistance) may all contribute, to some degree, to ddI resistance. L74V is the most common mutation during ddI monotherapy and causes between 2- and 8-fold ddI resistance depending on the genetic context in which it develops [St Clair CM, et al. 1991]. K65R causes 3- to 5-fold resistance to ddI but occurs rarely in clinical settings [Zhang D, et al. 1994; Winters MA, et al. 1997]. M184V confers 2- to 5-fold ddI resistance and occurs in about 10% of patients receiving ddI [Winters MA, et al. 1997]. V75T causes about 5-fold ddI resistance but has not been reported in patients receiving ddI monotherapy [Winters MA, et al. 1997].
Phenotypic resistance:
Drug resistance to ddI is difficult to detect in vitro. Conversion of ddI to its active form, i.e., ddATP, occurs best in resting cells and macrophages. Because nearly all current drug susceptibility assays take place in activated lymphocytes, the in vitro activity of ddI is blunted and loss of that activity may not always be detectable. In most assays, the dynamic range in susceptibility between wild-type isolates and the most highly drug-resistant isolates is approximately 10- to 20-fold. However, 2-fold decreases in susceptibility may indicate clinically significant drug resistance.
Q151M is associated with about 3-fold ddI resistance; the complex of Q151M together with V75I, F77L, and F116Y can cause high levels of ddI resistance (about 20-fold) [Lacey SF, et al. 1994; Iversen AK, et al. 1996]. During AZT/ddI combination therapy, low to moderate levels of ddI resistance (3-5 fold) often develop in an AZT-resistance context, i.e., in the presence of M41L, T215Y, other TAMs (thymidine analog mutations; aka nucleotide excision mutations or NEMs) [Japour AF, et al. 95; Holodniy M, et al. 1996]. Changes at codon 69 (which by itself causes measurable resistance to ddC but not to ddI) and noncanonical mutations (mutations that have not been proven to confer drug resistance, yet occur only in patients receiving antiretroviral therapy [i.e. they are not naturally occurring polymorphisms]) may facilitate low levels of ddI cross-resistance [Shafer RW, et al. 1996]. The b3-b4 insertion causes moderate levels of ddI resistance (3-5 fold), which are increased to higher levels in the presence of TAMs [Shafer RW, et al. 1998]. The presence of several TAMs together with other mutations, such as 118I and 69D, or L74V but not 184V, can also augment levels of resistance to ddI [Larder BA, Bloor S. 2001].
Cross-resistance:
The spectrum of mutations causing resistance to ddI overlaps with the mutations causing resistance to most other NRTIs. L74V causes resistance to ddC and abacavir. M184V causes resistance to 3TC, ddC and abacavir. K65R causes resistance to ddC, 3TC and adefovir. V75T causes resistance to d4T. Finally, multinucleoside resistance via Q151M or the b3-b4 insertion develops in about 10% of patients receiving NRTI combinations (see AZT resistance summary).
Although TAMs do not usualy cause detectable levels of ddI resistance on their own, there are three lines of evidence suggesting some degree of cross-resistance between AZT and ddI: (1) TAMs, in particular M41L and T215Y, develop in up to 10% of patients receiving ddI monotherapy [Winters MA, et al. 1998; Winters MA, et al. 1998a]; (2) M41L, T215Y and other TAMs develop significantly more rapidly in patients receiving AZT and ddI compared with AZT alone [Brun-Vezinet F, et al. 1997]; and (3) patients with HIV-1 isolates that have T215Y have markedly diminished virologic and immunologic responses to subsequent ddI therapy [Holodniy M, et al. 1996; Shafer RW, et al. 1999; D'Aquila RT, et al. 1995].
Emergence of resistance in vivo:
There are few published "complete" RT sequences (encompassing most of the polymerase coding region, codons 1-250) from patients receiving ddI monotherapy [Japour AF, et al. 1995]. Those that are available indicate that L74V develops in nearly two-thirds of patients treated for 1-2 years [Winters MA, et al. 1997]. M184V usually develops together with L74V in about 10% of treated patients, although M184V has been reported to occur alone in ddI-treated patients [Winters MA, et al. 1997]. The combination of M41L and T215Y occurs in about 10% of patients [Winters MA, et al. 1997; Japour AF, et al. 1995]. K65R and mutations at codon 69 rarely occur. During AZT/ddI combination therapy, L74V is almost completely prevented [Kojima E, et al. 1995; Larder BA, et al. 1996; Demeter LM, et al. 1995; shafer RW, et al. 1994; Shafer RW, et al. 1995; Coakley E, et al. 1999; Kozal MJ, et al. 1994; Lori F, et al. 1994; Gao WY, et al. 1994; Gao WY, et al. 1995; De Antoni A, et al. 1997]. Some patients develop dual resistance via the Q151M multinucleoside resistance pathway. Most patients appear to have virologic failure with HIV-1 isolates carrying multiple AZT resistance mutations together with other nonpolymorphic RT changes [Shafer RW, et al. 1996; Shafer Rw, et al. 1994]. During d4T + ddI therapy, escape viral mutants usually have classical TAMs and less commonly have the multinucleoside resistance mutations [Shafer RW, et al. 1995].
Clinical correlates of drug resistance:
In patients receiving ddI monotherapy, patients having isolates with L74V had an accelerated loss of CD4 counts compared with patients harboring wild-type isolates [Coakley E, et al. 1999]. In other settings, the correlation between ddI resistance and clinical outcome has been difficult to ascertain. First, phenotypic resistance is difficult to detect. Second, the genetic mechanisms of ddI resistance are altered during combination therapy and the full spectrum of genetic changes responsible for ddI resistance during combination therapy may not be known.
Comments: Potentiation by hydroxyurea (HU):
Hydroxyurea inhibits ribonucleoside reductase and interferes with HIV-1 DNA synthesis in lymphocytes and macrophages [Kozal MJ, et al. 1994]. Hydroxyurea inhibits the synthesis of dATP more than that of other intracellular nucleosides and thus potentiates ddI by reducing its competing substrate (dATP) and causing a compensatory increase in ddA triphosphorylation [Lori F, et al. 1994; Gao WY, et al. 1994]. Initial clinical trials have shown that the combination of ddI and hydroxyurea suppresses plasma HIV-1 RNA levels more than ddI alone. The mechanisms of resistance to the combination appear to be similar to the mechanisms of ddI resistance [Gao WY, et al. 1995]. However, the toxicity profile of HU has resulted in greatly diminished use of this compound in recent years.
Coakley, E., Gillis, J., and Hammer, S. Mutations in the reverse transcriptase genome of HIV-1 isolates derived from subjects treated with didanosine and stavudine in combination [abstract 116].
6th.Conf.Retrovir.Oppor.Infect., Chicago, IL, 1999.
Larder, B. A., and S. Bloor. Analysis of clinical isolates and site-directed mutants reveals the genetic determinants of didanosine resistance. Antivir. Ther. 2001;6:38.
Didanosine (ddI), Videx
