Xem mẫu

200 CHAPTER 15 the observation arm or the LDI treatment arm (3MU, subcutaneously t.i.w. for 3 years). Although a temporary effect on DFS was observed in the treat-ment arm, in the final analysis no DFS or OS benefit was observed in the WHO-16 study [13]. Another LDI (1MU, subcutaneously, on alternative days for 1 year) regimen was evaluated in the EORTC 18871/DKG-80 trial, showing not even a trend for a benefit [9]. Unfortunately, the impact on OS by HDI therapy was not confirmed by the recently unblinded ECOG 1690 study, in spite of a significant benefit on DFS [14]. Low-dose IFN treatment at 3MU for 2 years, in the ECOG 1690 trial did not demonstrate a benefit for either DFS or OS, just as in the WHO-16 trial. Conclusions on stage III Overall, it can be stated that observations have been inconsistent on the effi-cacy of IFN-a in the adjuvant setting for high-risk melanoma. Dose intensity as well as duration of treatment are not clearly defined and the efficacy of any regimen has yet to be demonstrated or confirmed by more than one trial. Table 15.2 summarizes the experience with IFN in adjuvant Phase III trials up to 1999. Trials in stage II In stage II patients with primary melanomas >1.5mm clinically node-negative, three trials in Europe have completed accrual. These three trials are similar in dosage, all using IFN-2aat low doses of 3MU for 6 months (Scottish trial), 12 months (Austrian trial) or 18 months (French trial). A preliminary report on the Scottish trial has not demonstrated a benefit in DFS or OS [1], whereas use for 12 months has been reported to result in a significant benefit on relapse rate [15]. The Austrian study has not reached maturity and so far no significant impact on OS has been observed. The French trial has reached ma-turity and a significantly prolonged DFS was observed in the IFN arm. The im-pact on OS failed to reach significance but demonstrated a favourable trend [16]. The NCCTG trial which evaluated the impact of HDI, intramuscularly t.i.w. for 3 months was negative both for DFS and OS in the stage II population of this mixed stage II–III trial [12]. Moreover, both ECOG 1684 and 1690 did not show a significant impact of HDI on DFS or OS in the stage IIB population of these trials [11–14]. In the ECOG 1690 trial in the LDI arm no effect of DFS or OS was observed on the stage IIB patients. Also in the EORTC/DKG-80 trial, the very low dose of 1MU, subcutaneously t.i.w. for 1 year, no impact on DFS or OS in the stage II patients in this trial was observed. So, in summary, we can state that in stage II melanoma in only two trials an impact of IFN on DFS was observed whereas in three other low-dose trials as well as in three high- THE ROLE OF BIOLOGICAL RESPONSE MODIFIERS 201 dose trials this has not been the case. A significant benefit on OS has not been reported in a single trial. Impact of sentinel node staging and stage migration The data on the impact of treatment with IFN are still unclear, especially whether the treatment will have any sizeble effect on OS. The EORTC-MCG has preferred to investigate a treatment option with considerable fewer side-effects in this stage II patient population which, as a consequence of the in-creasing use of sentinel node staging, will be transformed into a population with a much better prognosis and lower risk for relapse. Prognosis in sentinel node-negative patients is better than 95% survival at 4 years [17–19] and thus so excellent that it will no longer justify (even the evaluation of) toxic adjuvant therapy. Hence, in trial EORTC 18961 in a population of 1300 patients the efficacy of vaccination with the ganglioside vaccine GM2-KLH/QS-21 will be compared to the outcome in patients receiving standard of care (observation) [1]. Current and future trials The largest trial by far in high-risk melanoma patients (stage IIB–III) is the EORTC 18952 trial in 1300 patients. This trial evaluates the impact of inter-mediate doses of IFN where, after an induction period of 4 weeks, 5 days/week, 10MU, subcutaneously, is followed by a maintenance period of 10MU, subcutaneously t.i.w. for 1 year vs. 5MU subcutaneously t.i.w. for 2 years, vs. observation. This trial was designed to explore the effects of inter-mediate doses of IFN, hoping to identify a well-tolerated regimen that would have an ‘above threshold’ activity, comparable to the HDI schedule of ECOG 1684, but without the HDI-associated toxicity. The negative outcome of ECOG 1690 has somewhat undercut this philosophy. This trial will be analysed in the autumn of the year 2000. On the basis of the results with IFN in stage II patients and on the basis of the observation of a rebound in relapse rates in the IFN-treated patients in a number of trials (WHO-16 trial in stage III, French trial in stage II) the hypothesis has been raised that IFN needs to be administered for very long periods of time in order to be effective. This hypothesis is also based on the antiangiogenic mode of action of IFN, as demonstrated by Fidler and others [20,21]. The EORTC-MCG study will evaluate long-term therapy with IFN to standard of care (observation) in stage III melanoma. Long-term therapy has two prerequisites: low toxicity and easy administration. A well-tolerated dose 202 CHAPTER 15 of the pegylated form [22] of IFN-a (Peg-Intron) is being evaluated, as this agent needs only to be administered subcutaneously once a week, for a total treatment period of 5 years. This trial was activated in May 2001 (EORTC 18991). Roughly 50% of the total population of about 900 patients are ex-pected to enter the trial as patients with microscopic metastatic involvement of regional lymph node(s) as a consequence of the steady increase in sentinel node mapping in Europe. The other 50% will be patients with clinically overt (palpable) regional node involvement. Side studies regarding the value of re-verse transcription polymerase chain reaction (RT-PCR) of the sentinel node and other nodes in the regional node basin and of RT-PCR of blood samples will provide further insight into the biological importance and predictive value of such procedures. In this respect one must signal that the first report on the predictive value for relapse on the basis of RT-PCR on the sentinel node, by the group of Reintgen of the Lee Moffit Cancer Center, are very promising and convincing [19]. The ECOG is at present evaluating the GM2-KLH/QS-21 vaccine in stage IIB–III patients in the ECOG 1694 trial. In this trial 851 patients are random-ized into either HDI (ECOG 1684 schedule) or the vaccine arm. Early results have shown a survival benefit for HDI compared to vaccine but these results require longer follow-up. In stage II patients the EORTC 18961 trial will eval-uate in 1300 patients the impact of vaccination with GM2-KLH/QS-21 vs. standard of care (observation). The ECOG is evaluating in trial 1697 the im-pact of 4 weeks of IFN, 20MU/m2, intravenously, for 4 weeks vs. observation in 1420 patients. These huge numbers of patients are necessary because of the tremendous impact of sentinel node mapping and the excellent prognosis of sentinel node-negative patients. In the USA, two other very large adjuvant tri-als are ongoing. The Sunbelt trial evaluates the impact of sentinel node staging and the use of RT-PCR methods on sentinel node evaluation, and HDI (ECOG 1684 schedule) in a multiarm trial in 3000 patients [23]. Morton’s polyvalent melanoma cell vaccine (PMV), a melanoma cell line–based vaccine that thus far has only been studied in non-controlled trials [24,25] is now being evalu-ated in a large multicentre trial in 750 stage III patients. BCG+PMV treatment will be compared to treatment with BCG alone. Biological response modifiers as therapy for advanced melanoma Various IL-2 regimens and combinations with IFN-a have been tested in patients with advanced melanoma during the past decade. The response rate reported with IL-2 as a single agent or in combination with IFN-a is from 10 to 41% with a small, but remarkable proportion of long-term responders. Subsequently, regimens combining IL-2, IFN-aand chemotherapy (chemoim-munotherapy) have been evaluated in Phase II trials suggesting improved THE ROLE OF BIOLOGICAL RESPONSE MODIFIERS 203 response rates. Recent Phase III trials have investigated the role of chemoim-munotherapy for the treatment of advanced melanoma. Here the results of the clinical trials are discussed. Single agent and combination immunotherapy IL-2 as a single agent has yielded response rates of up to 24% in a number of Phase II trials [26–32]. The rationale of combining IL-2 and IFN-ais based on in vitro observations that IFN-a upregulates the expression of HLA class I molecules on tumour cells [33] and synergizes with IL-2 in activating immunological effector cells [34]. In larger Phase II studies (Table 15.3) [31,35–43], the objective response rates of the combination treatment of IL-2 and IFN-awas between 0 and 41%. However, the role of IFN-ain addition to IL-2 was never proven. The only randomized trial investigating IL-2 alone vs. IL-2+IFN-a was terminated because of low response rates in both treatment arms [31]. The published data concerning duration of responses and survival are too few to permit meaningful conclusions on long-term effects of IFN-a and IL-2 combinations. However, in most trials a proportion of patients is observed with durable remissions. In one cohort 17% of 65 patients survived for more than 3 years, and 11% for more than 5 years, of which 9% had no evidence of clinical disease at 5 years post treatment [37]. IL-2 has been tested in a variety of regimens. Initially, the maximal toler-able dose of IL-2 applicable under intensive care unit conditions was deter-mined to be 600000IU/kg every 8h for up to 5 days [35]. To investigate the role of treatment duration, two sequential studies were carried out with either a 5-day regimen of bolus IL-2 and IFN-aor a 3-day regimen. In this study, the 5-day regimen resulted in prohibitive cardiotoxicity and central nervous sys-tem toxicity. The 3-day regimen was associated with manageable toxicity. The response rate with the 5-day regimen was 41% and with the 3-day regimen only 20% [36]. West et al. [38] reported that continuous intravenous infusion of IL-2 is better tolerated than repeated bolus application and may be similarly effective. Addition of IFN-a to continuous intravenous infusion of IL-2 does not significantly increase toxicity and the response rates of this regimen range from 10 to 20% [38–40]. Based on the hypothesis that an initial high concentration of IL-2 is neces-sary to saturate the IL-2 receptors on non-activated T lymphocytes and a sub-sequent lower concentration may suffice to keep T lymphocytes activated, a decrescendo continuous infusion schedule has been developed. The IL-2 dose is increased fourfold as compared to the West protocol [38] over the initial 6h, and subsequently tapered to a low maintenence dose. In two sequential stud-ies, patients received 5 days of IFN-a followed either by continuous infusion of IL-2 at 18MU/m2/24h for 5 days (West protocol) or the decrescendo 204 CHAPTER 15 Table 15.3 Clinical trials of cytokine treatment in advanced melanoma Reference IL-2 only Rosenberg et al. [26] Thatcher et al. [27] Parkinson et al. [28] Whitehead et al. [29] Dorval et al. [30] Sparano et al. [31]* Legha et al. [32] IL-2±IFN-a Rosenberg et al. [42] Kruit et al. [39] Keilholz et al. [40] Sparano et al. [31]* Marincola et al. [43] Kruit et al. [36] EORTC-MCG [55]* IL-2±chemotherapy Stoter et al. [44] Dillman et al. [45] Flaherty et al. [46] Demchak et al. [47] Flaherty et al. [48] Atkins et al. [49] Dummer et al. [50] IL-2±IFN-a±chemotherapy Richards et al. [51] Khayat et al. [52] Buzaid & Legha [53] Proebstle et al. [54] EORTC-MCG [55]* IL-2 administration and concomitant treatment b. b. b. b. c.i.v. b. (reduced dose) c.i.v. b.+IFN-a c.i.v. (reduced dose)+IFN-a c.i.v.+IFN-a Decrescendo+IFN-a b. (reduced dose)+IFN-a b.+IFN-a b.+IFN-a(5 days) b.+IFN-a(3 days) Decrescendo+IFN-a c.i.v.+DTIC 850mg/m2 c.i.v.+LAK+DTIC 1,2g/m2 c.i.v.+DTIC 1g/m2 b.+CDDP 150mg/m2 s.c.+CDDP+DTIC b.+tamoxifen+CDDP+DTIC c.i.v.+DTIC 850mg/m2 b.+IFN-a+CDDP +DTIC+BCNU+tamoxifen c.i.v.+IFN-a+CDDP±tamoxifen b.+IFN-a+CDDP+DTIC Decrescendo+IFN-a+DTIC+CDDP Decrescendo+IFN-a+CDDP Number of patients 42 31 46 42 27 44 33 44 54 27 27 41 82 17 25 66 24 27 32 27 27 38 57 74 39 151 21 60 Response rate (%) 24 3 22 10 22 5 22 36 20 18 41 10 24 41 20 18 25 26 22 37 41 42 25 57 54 54 24 33 Duration (months) 2– >41 NR 4– >20 NR 4– >45 10– >14 NR NR NR 3–22 3– >36 2–15 NR 2–37 4– >10 3– >36 2–13 3– >24 2– >22 1– >30 3– >20 2– >20 NR 5– >10 3– >21 3– >24 4–9† 3– >36 Abbreviations: b., intravenous bolus; BCNU, carmustine; CDDP, cisplatin; c.i.v., continuous intravenous infusion; DTIC, dacarbazine; EORTC-MCG, European Organization for Research on Treatment of Cancer Melanoma Cooperative Group; IFN, interferon; LAK, lymphokine activated killer cells; s.c., subcutaneous. *Randomized study. †Very advanced disease patient population. ... - tailieumienphi.vn
nguon tai.lieu . vn