Within 1 year, 373 (48%) patients had a switch of the second regimen, with toxicity being the most common reason (n = 142) (Tables 1 and 3). Table 3 summarizes the incidence of the different reasons for switching second-line HAART, as well as the detectability of HIV-1 RNA (below or above 500 copies/ml) at the end of follow-up. The 1-year cumulative incidence of toxicity-driven switch on second-line HAART was 24% (95% CI, 21–28). The overall 1-year cumulative incidence of switching a second-line HAART regimen was 53%, indicating that approximately half of the switches was attributable to toxicity. Switching treatment for toxicity usually did not coincide with virological failure.
In a subgroup analysis among patients who had switched the first HAART regimen because of toxicity (n = 270), 77 (representing more than 10% of the current study cohort) subsequently had another toxicity-driven switch. The 1-year cumulative incidence of such subsequent toxicity-driven switches was 37% (95% CI, 30–44), indicating a higher risk of switching for toxicity for patients who had previously changed treatment likewise as a result of toxicity compared with patients who had not (Fig. 1).
After univariate analysis and stepwise multivariate analysis, four factors were identified that were associated with toxicity-driven switching of second-line HAART (Table 4). First, there was a risk reduction over calendar time. Second, women appeared to have a higher risk of toxicity-driven switches than men. Similarly, patients who switched the first HAART regimen because of toxicity had a 2.5-fold increased risk (95% CI, 1.7–3.5), relative to patients who switched because of other reasons, for toxicity-driven switch of the second regimen. Finally, among the various types of switch, switching from PI to nevirapine without changing the NRTI appeared to be protective against a subsequent switch for toxicity reasons (adjusted RR, 0.2; 95% CI, 0.1–0.6). Conversely, switching from PI to nevirapine with a concomitant change of NRTI was not favorable. Within the subgroup of patients who previously switched first HAART because of toxicity (Table 4, column 3), switching to nevirapine without changing the NRTI was equally favorable, whereas switching to nevirapine and concomitantly changing NRTI tended to increase the risk.
Body mass index and lean body mass as well as other baseline characteristics, such as nationality, route of HIV transmission, antiretroviral treatment experience prior to first HAART and duration of first HAART, did not significantly influence the incidence of toxicity-driven switch on the second regimen (data not shown). Stratification by hospital, in order to explore whether results may have been influenced by being under the care of a specific treatment centre, did not alter the conclusions (data not shown).
After having switched the first HAART because of gastrointestinal toxicity (n = 100), the 1-year risk of a recurrent toxicity-driven switch because of gastrointestinal toxicity was 26% (95% CI, 15–37), representing a 5.1-fold (95% CI, 2.1–12.1) increased risk relative to patients who switched the first HAART because of other toxicity reasons (n = 170). Out of 100 patients who switched the first HAART for reasons of gastrointestinal toxicity, 22 patients switched to nevirapine without changing NRTI. Only one of those had a recurrent switch because of gastrointestinal toxicity on the second regimen, suggesting a general improvement of gastrointestinal toxicity among patients who switched to nevirapine.
After having switched the first HAART because of neuropathy, the 1-year risk of another toxicity-driven switch because of neuropathy on the second regimen was 21% (95% CI, 5–37), yielding a RR of 12.8 (95% CI, 3.2–51.3) compared with patients who switched the first regimen because of other toxicity reasons. The small number of patients did not allow a distinction between the type of switch regarding the risk of neuropathy-driven switches on the second regimen.
Differentiating between those who concomitantly changed NRTI during the first switch and those who did not resulted in RR values for a recurrent switch because of neuropathy on the second regimen of 8.2 (95% CI, 1.5–45.1) and 19.2 (95% CI, 1.7–212.7) for these two groups of patients, respectively. These results suggest that changing the NRTI along with PI can reduce the risk of a neuropathy-driven switch. The fact that recovery from neuropathy is a time-consuming process might explain that the increased relative risk did not disappear.
In this cohort study, more than half of the patients who switched initial treatment for reasons other than virological failure again switched treatment within 1 year. Toxicity accounted for approximately half of these second switches and mostly involved gastrointestinal events and peripheral neuropathy. In line with previous reports, patients who switched a first HAART regimen for reasons of toxicity had a greater than twofold increased risk of a subsequent switch because of toxicity . This implies that a subgroup of patients may exist, including at least 10% of all patients, who are particularly prone to the development of HAART-associated toxicity. There was a slight reduction in risk of toxicity-driven switch over calendar time, which may reflect improved toxicity management as the experience of a physician evolves.
The proportion of toxicity-driven switches has previously been reported to be high for both first-line (26% within 1 year) [5,6] and second-line PI-containing HAART regimens (33% during a median of 483 days) . These findings, however, were based on the first regimen only and a relatively small single-centre patient population respectively. In a much larger and multicentre patient sample, we observed a 24% 1-year cumulative incidence of switching second-line regimens for toxicity. This figure may be slightly higher than for the entire population on second-line HAART, given that patients were selected who had a prior switch of treatment in the absence of virological failure and, therefore, may have been at potentially higher risk of a subsequent toxicity-driven switch. However, selecting such a population allowed us to investigate various types of switch, including switches between classes of antiretroviral drug, without the choice of subsequent regimens being influenced to any important extent by concerns of antiviral potency and cross-resistance between drugs.
An interesting observation was that switching to nevirapine, without concomitantly switching the NRTI, was associated with a fivefold lower risk of a subsequent toxicity-driven switch. The favourable effect of switching to nevirapine has been described before, albeit in small series of patients with a limited follow-up or a limited reference group [20–23]. Unfortunately, we were not able to study the effect of switching to other non-nucleoside analogue reverse transcriptase inhibitors, such as efavirenz and delavirdine, since these drugs were not licensed in the Netherlands during the period covered by the current analysis. Remarkably, combining the switch to nevirapine with a change in concomitant NRTI annihilated the advantageous effect of switching to nevirapine, possibly indicating the additional toxicity of the newly introduced NRTI. None of the other types of switch, including switching between different PI and changing NRTI at the same time, significantly influenced the risk of a subsequent toxicity-driven switch.
No associations were found between lean body mass or body mass index and the incidence of treatment switch for toxicity, suggesting little influence of distribution volume on toxicity risk for any of the drugs. Toxicity-driven switches occurred independently of patient nationality, which might suggest that cultural differences do not play a major role, but this association needs further substantiation. Women seemed to be more susceptible to toxicity. This may be attributable to sex-dependent pharmacokinetics or adherence leading to higher plasma concentrations of PI in particular . Alternatively, women may be more sensitive to toxicity or report toxicity more easily than men, resulting in a reporting bias.
Since this was an observational cohort study, results may have been confounded. For example, as guidelines for switching primarily focus on virological failure rather than toxicity [12,17,25], the decision to switch and how to switch was dependent of the treating physicians. They were also asked to indicate the primary reason for switching. As a consequence, the observed effects for the type of switch might in fact have been confounded by physician's preference. However, stratification by hospital did not alter the results, which to some extent reduces the likelihood of physician confounding. As we were able to adjust for the potential confounding effects of other determinants for toxicity-driven switches, the effect of residual confounding is assumed to be minimal.
In conclusion, toxicity continues to be a cause of major concern in the treatment of HIV-1 infection. It was deemed responsible for the majority of recurrent treatment switches among patients who had initially switched first-line HAART while having reasonable virus suppression, particularly among patients who had also switched their first-line HAART for toxicity. Switching to nevirapine rather than to an alternative PI, without concomitantly switching the NRTI, was associated with a significantly lower risk of again having to switch the new regimen for toxicity, even in patients with prior toxicity. These findings may provide guidelines for how to minimize the risk of renewed toxicity when managing patients with toxicity during PI-based HAART.
The authors thank the patients, physicians, nurses and data collectors participating in the ATHENA project and the ATHENA project team for their contribution to this study. We are grateful for the critical comments from the ATHENA Toxicity Working Group.
1. Vanhove GF, Schapiro JM, Winters MA, Merigan TC, Blaschke TF. Patient compliance and drug failure in protease inhibitor monotherapy. JAMA 1996, 276: 1955–1956.
2. Friedland GH, Williams A. Attaining higher goals in HIV treatment: the central importance of adherence. AIDS 1999, 13 (Suppl 1): S61–S72.
3. Flexner C. HIV-protease inhibitors. N Engl J Med 1998, 338: 1281–1292.
4. Deeks SG, Smith M, Holodniy M, Kahn JO. HIV-1 protease inhibitors. A review for clinicians.
JAMA 1997, 277: 145–153.
5. Ledergerber B, Egger M, Opravil M. et al
. Clinical progression and virological failure on highly active antiretroviral therapy in HIV-1 patients: a prospective cohort study. Lancet 1999, 353: 863–868.
6. d'Arminio Monforte A, Cozzi 1Lepri A, Rezza G. et al
. Insights into the reasons for discontinuation of the first highly active antiretroviral therapy (HAART) regimen in a cohort of antiretroviral naive patients. AIDS 2000, 14: 499–507.
7. van Roon EN, Verzijl JM, Juttmann JR, Lenderink AW, Blans MJ, Egberts AC. Incidence of discontinuation of highly active antiretroviral combination therapy (HAART) and its determinants. J Acquir Immune Defic Syndr Hum Retrovirol 1999, 20: 290–294.
8. Bini T, Testa L, Chiesa E. et al
. Outcome of a second-line protease inhibitor-containing regimen in patients failing or intolerant of a first highly active antiretroviral therapy. J AIDS 2000, 24: 115–122.
9. Mocroft A, Youle M, Moore A. et al
. Reasons for modification and discontinuation of antiretrovirals: results from a single treatment centre. AIDS 2001, 15: 185–194.
10. Carpenter CC, Cooper DA, Fischl MA. et al
. Antiretroviral therapy in adults. Updated recommendations of the International AIDS Society–USA Panel.
JAMA 2000, 283: 381–390.
11. Gazzard P, Moyle M on behalf of the BHIVA Guidelines Writing Committee. 1998 Revision to the British HIV Association guidelines for antiretroviral treatment of HIV seropositive individuals. Lancet 1998, 352: 314–316.
12. BHIVA Guidelines Co-ordinating Committee. British HIV Association guidelines for antiretroviral treatment of HIV seropositive individuals. Lancet 1997, 349: 1086–1092.
13. Lin JH, Chiba M, Chen IW, Nishime JA, Vastag KJ. Sex-dependent pharmacokinetics of indinavir: in vivo and in vitro evidence. Drug Metab Dispos 1996, 24: 1298–1306.
14. Gatti G, Di Biagio A, Casazza R. et al
. The relationship of ritonavir plasma levels and side effects: implications for therapeutic drug monitoring. AIDS 1999, 13: 2083–2089.
15. Gatti G, Vigano A, Sala N. et al
. Indinavir pharmacokinetics and parmacodynamics in children with human immunodeficiency virus infection. Antimicrob Agents Chemother 2000, 44: 752–755.
16. Dieleman JP, Gyssens IC, van der Ende ME, de Marie S, Burger DM. Urological complaints in relation to indinavir plasma concentrations in HIV-infected patients. AIDS 1999, 13: 473–478.
17. Borleffs JC, Danner SA, Lange JM, van Everdingen JJ. CBO guidelines: Antiretroviral therapy in the Netherlands. Ned Tijdschr Geneeskd 2001, 145: 1585–1589.
18. Gisolf EH, Jurriaans S, Pelgrom J. et al
. The effect of treatment intensification in HIV-infection: a study comparing treatment with ritonavir/saquinavir and ritonavir/saquinavir/stavudine. Prometheus Study Group.
AIDS 2000, 14: 405–413.
19. Council of State and Territorial Epidemiologists; AIDS Program, Center for Infectious Diseases. Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. MMWR 1987, 36 (Suppl 7): 1S-20S.
20. Dieleman JP, Gyssens IC, Sturkenboom MCJM, Niesters HGM, van der Ende ME. Substituting nevirapine for protease inhibitor because of intolerance. AIDS 1999, 13: 1423–1424.
21. De Luca A, Baldini F, Cingolani A. et al
. Benefits and risks of switching from protease inhibitors to nevirapine with stable background therapy in patients with low or undetectable viral load: a multicentre study. AIDS 2000, 14: 1655–1656.
22. Raffi F, Bonnet B, Ferre V. et al
. Substitution of a nonnucleoside reverse transcriptase inhibitor for a protease inhibitor in the treatment of patients with undetectable plasma human immunodeficiency virus type 1 RNA. Clin Infect Dis 2000, 31: 1274–1278.
23. Barreiro P, Soriano V, Blanco F, Casimiro C, de la Cruz JJ, Gonzalez-Lahoz J. Risks and benefits of replacing protease inhibitors by nevirapine in HIV-infected subjects under long-term successful triple combination therapy. AIDS 2000, 14: 807–812.
24. Burger DM, Koopmans PP, Brinkman K. Influence of gender on indinavir pharmacokinetics. Sixth European Conference on Clinical Aspects and Treatment of HIV Infection
. Hamburg 1997 [abstract 252].
25. Carpenter CC, Fischl MA, Hammer SM. et al
. Antiretroviral therapy for HIV infection in 1998: updated recommendations of the International AIDS Society–USA Panel. JAMA 1998, 280: 78–86.
Clinical and Epidemiological Working Group (*site coordinating physicians): W. Bronsveld, Medical Centre-Alkmaar; H. Weigel*, K. Brinkman and P. Frissen, Onze Lieve Vrouwe Gasthuis; J. ten Veen*, M. Hillebrand and S. Schieveld, Onze Lieve Vrouwe Gasthuis-Location Prinsengracht; J. Mulder*, E. van Gorp and P. Meenhorst, Slotervaart Hospital; A. van Eeden, J. van Goyen Kliniek; S. Danner*, F. Claessen* and R. Perenboom, Academic Hospital Vrije University; J. K. Eeftinck Schattenkerk, E. Gisolf, M. Godfried, J. van der Meer, J. Nellen, D. Notermans, K. Pogany, T. van der Poll, M. van Praag, J. Prins, P. Reiss, M. Reijers, Th. Ruys, M. van der Valk, A. Verbon and F. Wit, Academic Medical Centre, Amsterdam; C. Richter* and R. van Leusen Hospital Rijnstate, Arnhem; R. Vriesendorp, Westeinde Hospital, The Hague; R. Kauffmann* and E. Kogger, Hospital Leyenburg, The Hague; B. Bravenboer, Catharina Hospital, Eindhoven; C. ten Napel*, Medisch Spectrum Twente-Enschede; H. Sprenger* and G. Law, University Hospital, Groningen; R. W. ten Kate, Kennemer Gasthuis, Haarlem; M. Leemhuis, Medical Centre, Leeuwarden; F. Kroon* and E. Schippers, Leiden University Medical Centre; G. Schrey*, S. van der Geest and A. van der Ven, University Hospital, Maastricht; P. Koopmans*, M. Keuter and D. Telgt, University Hospital, Nijmegen; M. van der Ende*, I. Gyssens and S. de Marie, Erasmus University Medical Centre, Rotterdam; J. Juttmann* and C. van der Heul, St Elisabeth Hospital, Tilburg; M. Schneider*, J. Borleffs, I. Hoepelman and C. Jaspers, University Medical Centre, Utrecht; W. Blok, Hospital Walcheren,Vlissingen.
Subworking groups (*chair). Working Group on Cost-effectiveness: J. Tijssen*, G. Bonsel, M. Dijkgraaf and S. Heisterkamp, Academic Medical Centre, Amsterdam; Clinical Working Group: J. Lange*, M. Jambroes and G. J. Weverling, Academic Medical Centre, Amsterdam; Dutch HIV Patient Association: M. Mulder; Toxicity Working Group: J. Dieleman and I. Gyssens, Erasmus University Medical Centre, Rotterdam; K. Brinkman, Onze Lieve Vrouwe Gasthuis; P. Koopmans and H. ter Hoffstede, University Hospital, Nijmegen; P. Reiss*, G. J. Weverling and M. Jambroes, Academic Medical Centre, Amsterdam; Medical Psychology Working Group: M. Sprangers* and P. Nieuwkerk, Academic Medical Centre, Amsterdam; Pharmacology Working Group: D. Burger*, R. Aarnoutse and P. Hugen, University Hospital, Nijmegen; R. Hoetelmans*, R. van Heeswijk and A. Veldkamp, Slotervaart Hospital,Amsterdam; Virological Working Group: P. Rietra and K. Roozendaal, Onze Lieve Vrouwe Gasthuis; W. Pauw and A. van Zanten, Slotervaart Hospital; B. von Blomberg and P. Savelkoul, Academic Hospital Vrije University; F. de Wolf*, J. Goudsmit, L. van der Hoek and S. Jurriaans, Academic Medical Centre, Amsterdam; L. Nohlmans, Hospital Rijnstate, Arnhem; C. Jansen, Westeinde Hospital, The Hague; P. Franck and A. Lampe, Hospital Leyenburg, The Hague; E. Boel and A. Janz, Catharina Hospital, Eindhoven; R. Hendriks, Regional laboratory Twente, Enschede; J. Schirm, Regional laboratory, Groningen; H. Storm, Medical Centre, Leeuwarden; D. Veenendaal, LVF, Leeuwarden; A. Kroes*, Leiden University Medical Centre; C. Bruggeman and V. Goossens, University Hospital, Maastricht; J. Galama, University Hospital, Nijmegen; A. Osterhaus* and H. Niesters, Erasmus University Medical Centre, Rotterdam; A. Buiting, St Elisabeth Hospital, Tilburg; C. Boucher*, N. Back and R. Schuurman, University Medical Centre, Utrecht.
Steering Committee (*chair): J. Ruitenberg*, C. Boucher, D. Burger, S. Danner, R. Hoetelmans, R. W. ten Kate, R. Kauffmann, F. Kroes, J. Lange, A. Osterhaus, J. Tijssen and F. de Wolf.
Coordinating Centre: J. Lange, J. Tijssen, F. de Wolf (project leaders); M. Jambroes, E. van der Ven (project coordination); S. Brouwer, M. Overveld, R. van Boxtel (clinical research associates).
Data collection assistants: R. Runia and N. Wijdenes, Medical Centre, Alkmaar; N. Troost and R. Regez, Onze Lieve Vrouwe Gasthuis; M. Beerepoot, Onze Lieve Vrouwe Gasthuis-Location Prinsengracht; E. Oudmaijer, Slotervaart Hospital; J. Troon, Jan v. Goyen Kliniek; A. van Diggelen, Academic Hospital Vrije University; J. Ruijs and L. Veenenberg, Academic Medical Centre, Amsterdam; N. Langebeek, Hospital Rijnstate, Arnhem; M. Groot and S. Wildebeest, Westeinde Hospital, The Hague; A. de Haas, Hospital Leyenburg, The Hague; W. van Schaik and N. Slegers, Catharina Hospital, Eindhoven; H. Heins and T. Lansink, Medisch Spectrum Twente, Enschede; A. Bakker and S. Moolenburgh, University Hospital, Groningen; E. Kloosterhuis and M. Schoemaker, Kennemer Gasthuis, Haarlem; J. de Groot and A. Ketser, Medical Centre, Leeuwarden; W. Dorama, Leiden University Medical Centre; Ch. Leenders, University Hospital, Maastricht; M. Meeuwissen and B. Zomer, University Hospital, Nijmegen; T. Royaards, Erasmus University Medical Centre, Rotterdam; R. Santegoets and B. van der Ven, St Elisabeth Hospital, Tilburg; F. Bär, University Medical Centre, Utrecht; S. Baas and C. Ruissen, Hospital Walcheren, Vlissingen.