Reversible myelosuppression is a major side effect of chemotherapy that predisposes to opportunistic infections even in immunocompetent patients. The recovery of myeloid lineage cells occurs within 14–21 days by replenishment from haematopoetic stem cells. Lymphoid cells in contrast, include both long-lived non-cycling naive cells and shorter-lived effector and memory cells and their recovery after chemotherapy is slower, more variable and less complete . Chemotherapy-induced immunodeficiency is characterized by T-lymphocyte depletion, although falls in B cells and natural killer cells accompany this [2–4]. Chemotherapy in immunocompetent patients causes a more profound decline in CD4 than CD8 T cells  whereas the natural killer cell population is relatively spared .
The most striking finding from these studies in immunocompetent patients is the protracted recovery of CD4 cells. Unlike CD8 cells, B cells and natural killer cells, that all recover within 3 months of completing chemotherapy [7–9], the CD4 cell count is only one-third of the pre-treatment level at 3 months after completing treatment and may not have recovered to pre-treatment levels after 1 year . HIV-1 affects immune function primarily by infection and destruction of CD4 cells . There is therefore concern that prolonged CD4 suppression by chemotherapy may have a major adverse influence on the course of HIV-1 disease.
The last 4 years have seen a revolution in the treatment of HIV-1 with the introduction of highly active antiretroviral therapy (HAART) which suppresses plasma HIV-1 RNA load and increases CD4 T-lymphocyte count . Little is known about the immunological effect of chemotherapy in individuals receiving HAART. It was hoped that the addition of HAART would have a beneficial effect on the incidence of opportunistic infections (OI) and indirectly the tumour response rate by enabling patients to receive their chemotherapy without delays and dose modifications. Unfortunately, early trials failed to demonstrate these advantages [12,13]. Indeed current trials at the National Cancer Institute advocate stopping HAART for the duration of chemotherapy . In order to improve our understanding of how HAART and chemotherapy interact in relation to immune function, we have prospectively measured immune parameters in patients receiving chemotherapy for AIDS-related lymphoma (ARL) while continuing on HAART. This study has important implications for opportunistic infection prophylaxis particularly against Pneumocystis carinii pneumonia (PCP) and Mycobacterium avium complex (MAC).
Patients and methods
A prospective study enrolled HIV-1 seropositive patients who were receiving combination chemotherapy for AIDS-related non-Hodgkin's lymphoma (NHL) or Hodgkin's disease (HD) and concomitant HAART. This was defined as at least three antiretroviral agents including at least one protease inhibitor (PI) or non-nucleoside reverse transcriptase inhibitor (NNRTI). The peripheral blood total lymphocyte, CD4, CD8, CD19 (B cell), CD16 and CD56 natural killer cell counts and plasma HIV RNA load were measured at lymphoma diagnosis, at 1 and 3 months during chemotherapy and 1, 3 and 6 months after the completion of chemotherapy.
During the study period (April 1996–December 1999) ARL was diagnosed in 61 patients in our unit. Forty-seven patients had systemic NHL, nine had primary cerebral NHL and five had HD. Twenty-nine of these patients (five HD and 24 systemic NHL) were treated with curative intent and received concomitant chemotherapy and HAART. Complete immunological data including measurements following completion of chemotherapy is available for 20 of these 29 patients. The data is incomplete for the remaining patients on account of early deaths (n = 3), chemotherapy stopped at patients request (n = 3) and patients completing their treatment elsewhere (n = 3).
Total lymphocyte and subset analysis was performed using whole blood stained with murine anti-human monoclonal antibodies to CD3, CD4, CD8, CD16, CD19 and CD56 (TetraOne; Beckman Coulter, High Wycombe, UK) and were evaluated on an Epics XL-MCL (Beckman Coulter) flow cytometer.
Plasma viral RNA assay
Viral load in patient plasma was measured at each time point of sample collection using the Quantiplex HIV RNA 3.0 (Chiron bDNA) assay (detection limit < 50 HIV-1 copies/ml; Chiron Diagnostics, Halstead, UK).
The mean age of the cohort of 20 patients was 42.2 years (range, 22–57 years) and all but one patient were male. Two patients had HD and 18 systemic NHL. The chemotherapy treatments were infusional cyclophosphamide, doxorubicin and etoposide (CDE)  for 10 patients, bleomycin, etoposide, methotrexate, vincristine, prednisolone, cyclophosphamide and doxorubicin (BEMOP/CA)  for nine patients (including one with HD) and doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD)  for one patient with HD. The duration of chemotherapy was 3–6 months. The clinicopathological details are shown in Table 1.
The concurrent antiretroviral therapy was based upon previous antiretroviral therapy and current unit practice. The HAART is therefore not uniform for all patients but represents a spectrum of common combinations used by HIV-1-infected individuals. Nine patients were receiving a PI-based regimen, eight patients an NNRTI-based regimen and three patients were being treated with a schedule including both a PI and a NNRTI (see Table 1). Fourteen patients had received prior antiretroviral therapy regimens and the median number of prior regimens was two (range, 0–10). Six patients had previously received a PI, two patients an NNRTI and four patients both drug classes.
No alterations to antiretroviral therapy were made during the study period regardless of the immunological results. In addition patients received PCP and MAC prophylaxis during chemotherapy regardless of their CD4 T-cell count at entry, using co-trimoxazole or pentamidine and azithromycin or isoniazid. The median follow-up was 1.2 years (maximum, 4 years) and six patients have died, five from lymphoma and one from neutropenic sepsis. The major toxicities were haematological. In addition to the one fatal neutropenia, seven patients (35%) developed WHO grade 4 neutropenia, three (15%) developed grade 4 thrombocytopenia and one developed (5%) grade 4 anaemia.
In this selected cohort, the complete response rate was 65% and the partial response rate was 20%; the 1-year overall survival was 79% (95% confidence interval, 61–97%). There were no significant differences in overall survival between groups who received PI-based, NNRTI-based or dual HAART (log-rank test, P = 0.70).
The T-helper cell (CD4) count fell significantly from a pre-treatment median of 138 × 106 cells/l [interquartile range (IQR), 174 × 106 cells/l] to a median of 77 × 106 cells/l (IQR, 83 × 106 cells/l) after 3 months of chemotherapy (paired t-test, P = 0.0019). After treatment was completed the CD4 T-cell count increased rapidly to a median of 139 × 106 cells/l (IQR, 149 × 106 cells/l) at 1 month and 187 × 106 cells/l (IQR, 166 × 106 cells/l) at 3 months. Within 1 month of completing the chemotherapy the CD4 T-cell count was not significantly different from pre-treatment levels (paired t-test, P = 0.55) (see Fig. 1).
The T-cytotoxic cell (CD8) count also fell from a median of 867 × 106 cells/l (IQR, 983 × 106 cells/l) before treatment to 293 × 106 cells/l (IQR, 333 × 106 cells/l) after 3 months of chemotherapy, although this was not a statistically significant difference (paired t-test, P = 0.09). Although the recovery in the CD8 T-cell count was prolonged – at 3 months following completion of treatment the median was 505 × 106 cells/l (IQR, 708 × 106 cells/l) and at 6 months it was 749 × 106 cells/l (IQR, 588 × 106 cells/l) – none of these values were significantly different from pre-treatment values (see Fig. 1).
There was no significant alteration in the CD4 T-cell percentage during the study period (paired t-test, P = 0.64) and this suggests that the falls in CD4 cell count reflected a global decline in lymphocytes during chemotherapy and their recovery after chemotherapy. This is supported by the profile of CD3 T cells which fell from a median of 1138 × 106 cells/l (IQR, 1072 × 106 cells/l) to a nadir after 3 months of chemotherapy of 400 × 106 cells/l (IQR, 316 × 106 cells/l) (paired t-test, P = 0.002). The mean CD3 cell count had recovered within 1 month of ending the chemotherapy to pre-treatment levels (paired t-test, P = 0.06).
The natural killer (CD16/CD56) cell count similarly fell significantly from a pre-treatment median of 139 × 106 cells/l (IQR, 78 × 106 cells/l) to a nadir after 3 months chemotherapy of 74 × 106 cells/l (IQR, 75 × 106 cells/l) (paired t-test, P = 0.008). These cells also recovered promptly within 1 month of the end of chemotherapy (paired t-test, P = 0.054) (see Fig. 1).
The B lymphocyte count (CD19 cells) also fell from a pre-treatment median of 125 × 106 cells/l (IQR, 157 × 106 cells/l) to a nadir of 12 × 106 cells/l (IQR, 30 × 106 cells/l) after 3 months of chemotherapy (paired t-test, P = 0.0033). The recovery following chemotherapy was slower with levels at 1 month that were still significantly lower than pre-treatment levels (28 × 106 cells/l (IQR, 56 × 106 cells/l) (paired t-test, P = 0.0084). However, by 3 months after the end of chemotherapy the CD19 count was not significantly lower than the pre-treatment level (paired t-test, P = 0.09) (see Fig. 1).
There was no significant change in the HIV-1 RNA load during the whole study period, either during the chemotherapy or following completion of chemotherapy (see Fig. 1). Prior to chemotherapy the median HIV-1 RNA load was 165 copies/ml and 37% had viral loads below the level of detection of the assay (< 50 copies/ml). After 3 months of chemotherapy the median was 600 copies/ml and 32% had undetectable levels. At 3 months after completing chemotherapy the median viral load was 190 copies/ml and 35% had undetectable levels.
There were no differences in lymphocyte subset counts or plasma HIV-1 load between patients with NHL and HD or between the different chemotherapy or HAART regimens.
The assessment of immune function in patients during and following chemotherapy for ARL has been difficult because the effect of the chemotherapy on immune cell counts and viral load was uncertain with limited published data derived mainly from patients who were not receiving concomitant HAART. These data describe the immunological profile of patients with ARL who were treated with concomitant chemotherapy and HAART. The CD4 T-lymphocyte count falls by approximately 50% but recovers rapidly within 1 month after treatment, and reflects a global fall in T cells, as the percentage of CD4 T cells remains unchanged throughout. This recovery is faster than predicted from HIV-1 seronegative patients or from patients with ARL who were treated with chemotherapy alone. In one study, six patients with ARL were treated with chemotherapy without concomitant antiretroviral drugs. After three cycles of chemotherapy, the CD4 T-cell count had halved and remained suppressed at 4 months after completing chemotherapy, compared with pre-treatment levels. Meanwhile, the viral load rose significantly from 0.6 to 2 logs during the first 9 weeks of therapy but quickly returned to pre-treatment levels after the completion of chemotherapy. In contrast, the CD8 T-cell count did not fall by a significant amount, which concurs with previous reports .
In immunocompetent patients receiving chemotherapy, the CD4 T-cell count may be suppressed for more than 1 year following chemotherapy whereas the CD8 T-cell count recovers more quickly . These comparisons suggest that the rapid recovery of CD4 T-cell count in our cohort is due to the continuing treatment with HAART. We postulate that the constant thymic stimulation caused by HAART  is responsible for the more rapid rise in CD4 T-cell count following chemotherapy. As expected, the immunocompetent HIV-1 seronegative patients that have been studied have a higher CD4 T-cell count before chemotherapy , but demonstrate a dramatic fall in CD4 T cells during chemotherapy. It is perhaps unsurprising that the CD4 T-cell recovery to this initially higher level takes longer than that observed for the HIV-1 population on HAART who have active T-cell regeneration. It would be interesting to see if, as speculated, there is a higher proportion of CD45RA CD4 T cells in the HAART patients after chemotherapy compared with the HIV-1 seronegative population, which may rely on thymic independent pathways for T-cell regeneration [5,20].
The use of HAART during chemotherapy has a number of potential complications including pharmacokinetic interactions and overlapping toxicities that might compromise both the lymphoma treatment and long-term HIV-1 control. The National Cancer Institute has adopted a dose-adjusted chemotherapy schedule for the management of ARL and antiretroviral therapy is suspended for the duration of the chemotherapy. Preliminary results have been reported for 30 patients with an overall survival at 30 months of 74%. Plasma HIV-1 RNA load control was re-established within 3 months of completing chemotherapy and restarting HAART but the CD4 T-cell count took 12 months to recover to baseline values .
Studies in HIV-1-seropositive individuals have shown that without HAART the viral load increases during chemotherapy [14,18]. The concomitant use of chemotherapy and HAART in our study resulted in no change in viral load over the same time period. The ability to estimate the nadir CD4 T-cell count during chemotherapy has implications for prophylaxis against PCP and MAC infections, in which the risk is related to the absolute level of CD4 T cells. Following these results we have adopted a policy of prescribing PCP prophylaxis if levels are predicted to fall below 200 × 106 cells/l and MAC prophylaxis if levels are predicted to fall below 50 × 106 cells/l.
In conclusion, our results show that it is feasible to continue HAART during chemotherapy for AIDS-related lymphoma. With this policy, immune function is maintained as well as possible during chemotherapy when patients are at high risk of infection. The results of this study will help to identify patients who will require opportunistic infection prophylaxis during chemotherapy.
1. Mackall C. T-cell immunodeficiency following cytotoxic antineoplastic therapy: A review. Stem Cells 2000, 18: 10–18.
2. Esber E, DiNicola W, Movassaghi N, Leiken S. T and B lymphocytes in leukaemia therapy. Am J Hematol 1976, 1: 211–218.
3. Mackall CL, Fleisher TA, Brown MR, Magrath IT, Shad AT. Lymphocyte depletion during treatment with intensive chemotherapy for cancer. Blood 1994, 84: 2221–2228.
4. Cheson B. Infectious and immunosuppressive complications of purine analog therapy
. J Clin Oncol 1995, 13: 2431–2448.
5. Hakim FT, Cepeda R, Kaimei S, Mackall CL, McAtee N. Constraints on CD4 recovery post chemotherapy in adults. Blood 1997, 90: 3789–3798.
6. Komada Y, Zhand SL, Zhou YW. et al
. Cellular immunosuppression in children with acute lymphoblastic leukaemia: effect of consolidation chemotherapy. Cancer Immunol Immunother 1992, 35: 271–276.
7. Mackall CL, Fleisher TA, Brown MR, Andrich MP, Chen CC. Distinctions between CD8 and CD4 regenerative pathways result in prolonged T cell subset imbalance after intensive chemotherapy. Blood 1997, 89: 3700–3707.
8. Small T, Keever C, Weiner-Fedus S, Heller G, O'Reilly R, Flomenberg N. B cell differentiation following autologous, conventional or T cell depleted bone marrow transplantation: a recapitulation of normal B cell ontogeny. Blood 1990, 76: 1647–1656.
9. Storek J, Saxon A. Reconstitution of B cell immunity following bone marrow transplantation. Bone Marrow Transplant 1992, 9: 395–408.
10. Haase A. Population biology of HIV-1 infection: viral and CD4+ T cell demographics and dynamics in lymphatic tissues. Annu Rev Immunol 1999, 17: 625–656.
11. Autran B, Carcelian G, Li S, Blanc C, Mathez D. Positive effects of combined antiretroviral therapy on CD4 Tcell homeostasis and function in advanced HIV disease. Science 1997, 227: 122–126.
12. Sparano JA, Wiernik PH, Hu X, Sarta H, Henry DH, Ratech H. Saquinavir enhances the mucosal toxicity of infusional cyclophosphamide, doxorubicin and etoposide in patients with HIV-associated non-Hodgkin's lymphoma. Med Oncol 1998, 15: 50–57.
13. Vaccher E, Spina M, Santarossa S, et al
. Concomitant CHOP chemotherapy and highly active antiretroviral therapy (HAART) in patients with HIV related non-Hodgkin's lymphoma. XII International Conference on AIDS
. Geneva, June 1998 [abstract 22289].
14. Little R, Pearson D, Gutierrez M, Steinberg S, Yarchoan R, Wilson W. Dose-adjusted chemotherapy with suspension of antiretroviral therapy for HIV-associated non-Hodgkin's lymphoma. J Acquir Immune Defic Syndr 2000, 23: A11.A11.
15. Sparano JA, Wiernik PH, Hu X. et al
. Pilot trial of infusional cyclophosphamide, doxorubicin and etoposide plus didanosine and filgrastim in patients with HIV associated non-Hodgkin's lymphoma. J Clin Oncol 1996, 14: 3026–3035.
16. Bower M, Stern S, Fife K, Nelson M, Gazzard BG. Weekly alternating combination chemotherapy for good prognosis AIDS related lymphoma. Eur J Cancer 2000, 36: 363–367.
17. Canellos G, Anderson J, Propert K. et al
. Chemotherapy of advanced Hodgkin's disease with MOPP, ABVD or MOPP alternating with ABVD. N Engl J Med 1992, 327: 1478–1484.
18. Zanussi S, Simonelli C, D'Andrea M. et al
. The effects of antineoplastic chemotherapy on HIV disease. AIDS Res Hum Retrovir 1996, 12: 1703–1707.
19. Douek DC, McFarland RD, Keiser PH. et al
. Changes in thymic function with age and during the treatment of HIV infection. Nature 1998, 396: 690–695.
20. Haynes BF, Markert ML, Sempowski GD, Patel DD, Hale LP. The role of the thymus in immune reconstitution in aging, bone marrow transplantation, and HIV-1 infection. Annu Rev Immunol 2000, 18: 529–560.