Docetaxel/Gemcitabine: Salvage Chemotherapy in Anthracycline-Pretreated Patients With Advanced Breast Cancer

February 1, 2001

Docetaxel (Taxotere) and gemcitabine (Gemzar) are active agents against breast cancer. Several phase I studies evaluated different schedules of their combination and clearly demonstrated that docetaxel and gemcitabine can

ABSTRACT: Docetaxel (Taxotere) and gemcitabine (Gemzar) are active agentsagainst breast cancer. Several phase I studies evaluated different schedules oftheir combination and clearly demonstrated that docetaxel and gemcitabine can besafely combined in either an every 3-week schedule or in a weekly and biweeklyschedule. The toxicity profiles of these combinations were mainly grade 3 and 4neutropenia and asthenia. Phase I studies also suggested that thedocetaxel/gemcitabine combinations are active regimens in pretreated patientswith advanced breast cancer. Three phase II studies of patients previouslytreated with anthracycline-based chemotherapy reported a mean objective responserate of 46% and a mean overall survival of 13.5 months. Two of these trialsenrolled patients with anthracycline-resistant or anthracycline-refractorydisease; the objective responses using docetaxel/gemcitabine combination were36% to 55% and 54%, respectively. It is noteworthy that objective responses werealso achieved with this regimen in some patients who progressed while receivingtaxane-based, front-line chemotherapy. These efficacy results were obtained witha mild toxicity profile. Adverse events were of short duration and easilymanageable. Further studies are needed to evaluate this combination asfront-line chemotherapy as well as second-line in well-defined subgroups ofpatients with advanced breast cancer. Furthermore, the combination should becompared with other more standard or investigational regimens. [ONCOLOGY15(Suppl 3):18-24, 2001]

Introduction

Since metastatic breast cancer remains incurable despitetemporary regression of the disease with endocrine therapy or chemotherapy,palliation of symptoms and prolongation of high-quality life become the majortherapeutic goals in the treatment of these patients. Combination chemotherapyused as front-line treatment for metastatic disease usually results in 35% to75% objective responses. However, the complete response rate is relatively low(approximately 10%) and the average duration of response only 8 months.[1]Moreover, when disease progression occurs, a standard second-line combinationchemotherapy produces responses in 20% to 45% of patients, depending on whetheran anthracycline was included in the initial treatment.[2,3]

Rationale

Docetaxel

Docetaxel (Taxotere) is a new semisynthetic taxane withsignificant anti-neoplastic activity and manageable toxicity, which consists primarily ofmyelosuppression. As front-line monotherapy for metastatic breast cancer, it produced an overall response rate of59%; as second-line therapy it achieved 46%.[4] Even for patients who relapsedor progressed after prior anthracycline-based chemotherapy, the objectiveresponse rate was 41%.[5]

In addition, a randomized phase III study evaluateddocetaxel/doxorubicin versus doxorubicin/cyclophosphamide (Cytoxan, Neosar) asfront-line chemotherapy of patients with metastatic breast cancer. The studydemonstrated a significantly higher objective response rate (ORR) withdocetaxel/doxorubicin (ORR = 60%) than with doxorubicin/cyclophosphamide (ORR =47%; P = .008).[6] Docetaxel has a unique mechanism of action; it disruptsmitosis, promotes microtubular assembly, and suppresses depolymerization ofmicrotubular bundles to free tubulin.[7]

Gemcitabine

Gemcitabine (Gemzar) is a novelS-phase-specific, fluorine-substituted pyrimidine analog, phosphorylated bydeoxycytidine kinase to the active diphosphate and triphosphate metabolites. Gemcitabine diphosphate inhibitsribonucleotide reductase, which results in the depletion of the intracellulardeoxycytidine triphosphate pools. Reduced deoxycytidine triphosphate levelspromote the competitive incorporation of gemcitabine triphosphate into DNA,which results in the inhibition of further DNA synthesis by masked-chaintermination.[8]

Gemcitabine has a broad range of activity against varioustumors[9] and an especially favorable toxicity profile of mild myelosuppressionand minimal nonhematologic toxicity.[10] When used as a single agent, responserates ranged from 25% to 46%, depending on the dose administered and whetherpatients had previously received chemotherapy.[11]

Combination Regimen

The distinct mechanisms of action, different intracellulartargets, and activity of both docetaxel and gemcitabine against various tumors provided the rationale for their combination.Provided that pharmacologic antagonism does not occur and that clinicallyrelevant doses of both drugs can be delivered together, the combination offersthe potential for superior antitumor activity compared with the individualdrugs.

This review presents data from phase I and II studies of thedocetaxel/gemcitabine combination in patients with advanced breast cancer.

Phase I Studies

Weekly Gemcitabine andMonthly Docetaxel

Several phase I studies have evaluated the tolerance ofdifferent schedules of the docetaxel/gemcitabine combination in advanced breastcancer.

The first study of the combination by Spiridonidis et al[12]administered docetaxel at escalated doses (45, 60, 75, and 100 mg/m2) either onday 1 or day 15 while gemcitabine was given in a fixed dose (800 mg/m2) on days1, 8, and 15. The cycles were repeated every 4 weeks. In all, 40 patients wereenrolled and 132 chemotherapy cycles were delivered with a mean of threecycles/patient (range: 1 to 11 cycles).

There were 27 and 13 patients enrolled in the day-1 and day-15docetaxel schedules, respectively. In the day-15 schedule with docetaxel at doselevels of 45, 60, and 75 mg/m2, only 6 of the 13 patients were able to receiveall of the planned gemcitabine administrations during the first cycle as perprotocol. Overall, the day-15 dose was omitted in 9 of 35 docetaxel cyclesbecause of thrombocytopenia and elevated liver enzymes, which occurred duringthe first cycle in seven patients. This excessive toxicity resulted in adecreased dose intensity of the drugs in the day-15 docetaxel schedule comparedwith the day-1 schedule, enabling the investigators to prematurely close thisarm of the trial.

Conversely, using the day-1 schedule, docetaxel could beescalated to the dose of 100 mg/m2, which is the maximum tolerated dose forsingle-agent docetaxel.[10,11] In the day-1 docetaxel schedule, the day-8gemcitabine dose was omitted in 19 cycles because of neutropenia (18 cycles) andthrombocytopenia (1 cycle). During the day-1 docetaxel schedule, the dose-limiting toxicity was reached at the dose of docetaxel100 mg/m2 because of two episodes of febrile neutropenia and one of livertoxicity. Nonhematologic toxicity (asthenia) occurred in 30 patients; it wasgraded as moderate to severe in 15 patients. The onset of asthenia wasunpredictable, and its severity did not appear to be cumulative. The fluidretention syndrome was relatively rare (only three patients presented withsevere pedal edema). Grade 3 liver toxicity and grade 3 mucositis occurred intwo patients each.

Different/Escalated Doses

Rischin et al[14] conducted a phase I trial evaluating differentdoses of gemcitabine (800, 1,000, and 1,200 mg/m2) given on days 1 and 8 incombination with escalated doses of docetaxel (60, 75, 85, and 100 mg/m2) givenon day 8 in cycles of 21 days. The study enrolled 39 previously treated patientswho received a total of 152 chemotherapy cycles (median: 4 cycles/patient;range: 1 to 8 cycles).

Dose-limiting toxicity was reached at docetaxel 100 mg/m2 andgemcitabine 1,200 mg/m2, since three of six enrolled patients developed dose-limitingtoxicity events. One patient each experienced febrile neutropenia plus grade 3asthenia, febrile neutropenia plus grade 3 mucositis, and febrile neutropeniaplus grade 4 neutropenia lasting more than 7 days plus grade 4 thrombocytopenia.Neutropenia was the predominant hematologic toxicity, with grade 4 neutropeniaoccurring in 64% of patients; febrile neutropenia occurred in nine patients(seven in the first cycle) who experienced 11 episodes. Grade 4 neutropenialasting for more than 7 days occurred in 14 cycles. Grade 3 or 4 nonhematologictoxicity or thrombocytopenia was infrequent, with only one patient developing grade 4 skintoxicity.

The recommended doses for further phase II studies were definedas docetaxel 85 mg/m2 and gemcitabine 1,200mg/m2 every 21 days. However, atthis recommended level, 4 of 10 patients required treatment delay for a total of11 out of 30 cycles because of slow neutrophil recovery.

Phase II Studies

It is interesting to note that in all of these phase I trialsobjective responses were documented in patients with head and neck cancer, non-small-celllung cancer, and metastatic breast cancer. Therefore, it was reasonable tofurther evaluate the activity of the docetaxel/gemcitabine combination inpatients with advanced breast cancer. To the best of our knowledge, no phase IItrials have evaluated this combination in chemotherapy-naive patients withadvanced breast cancer. Conversely, three phase II studies evaluating theefficacy and tolerance of docetaxel/gemcitabine have already been conducted inpatients who were pretreated with anthracycline- and/or taxane-based front-linechemotherapy regimens (Table 2).

Efficacy Results

In the first study, conducted by the Greek Cooperative Group forBreast Cancer (Mavroulis et al), 52 patients with advanced breast cancerreceived as second-line chemotherapy docetaxel 100 mg/m2 on day 8 andgemcitabine 900 mg/m2/wk on days 1 and 8 in cycles of 21 days. From days 9 to15, rhG-CSF was given prophylactically to all patients.[24] In all, 25 (48%)patients had disease not responding to initial chemotherapy (anthracyclinerefractory) and 27 (52%) had disease relapsing after an initial response tofront-line chemotherapy (anthracycline resistant). In addition, 25 patients hadalso received taxane-based chemotherapy. The median interval between the end ofthe first-line treatment and the initiation of the docetaxel/gemcitabineregimens was 6 months (range: 1 to 60).

Overall, complete response was achieved in 7 (14%) patients andpartial response in 21 (40%), for an overall response rate of 54% (95%confidence interval [CI]: 40%-67%). Of 25 patients who had previously receiveda taxane-based front-line regimen, 11 responded to treatment (44%, 1 completeand 10 partial responses). It is interesting that 4 of the 11 responders haddisease progression under front-line taxane-based chemotherapy (3 of these 4patients had received docetaxel-based chemotherapy). The median duration ofresponse was 3.6 months (range: 1 to 16 months), the median time to tumorprogression was 8 months (range: 2 to 18.5 months), and the overall mediansurvival was 14 months (range: 1 to 16 months). Grade 3 and 4 neutropeniaoccurred in 10 and 5 (overall, 29%) patients, respectively. Febrile neutropeniawas observed in 4 patients. Grade 3 and 4 thrombocytopenia, occurred in 9 and 2patients, respectively.

The second phase II trial was conducted by the HellenicCooperative Oncology Group (Fountzilas et al) in 39 evaluableanthracycline-resistant patients with advanced breast cancer who had previousexposure to an anthracycline or mitoxantrone (Novantrone) either in the adjuvantsetting (providing that they had relapsed within 12 months of completion ofchemotherapy) or for advanced disease (and subsequently progressed).[25]Gemcitabine was given at the dose of 1,000 mg/m2 on days 1 and 8 followed bydocetaxel 75 mg/m2 on day 1 in cycles of 21 days.[17]

Three (7.5%) complete and 11 (28%) partial responses wereachieved for an overall response rate of 36% (95% CI: 21%-53%). The medianduration of response was 10.3 months (range: 4.6 to 17.5 months) and the mediantime to tumor progression was 7 months. The overall median survival was 12.7months. Grade 3 and 4 neutropenia occurred in almost half of the patients andwas not cumulative. Seven patients were hospitalized for febrile neutropenia. Atotal of 35 (90%) patients required rhG-CSF administration.

In the thrid study, Laufman et al[26] treated 39 evaluablepatients with docetaxel 100 mg/m2 on day 1 and gemcitabine 800mg/m2 on days 1,8, and 15 in cycles of 28 days without rhG-CSF support. However, only a smallportion of the enrolled patients had anthracycline-resistant disease. There were3 (7.7%) complete and 16 (41%) partial responses achieved for an overallresponse rate of 59% (95% CI: 41%-76%). The median time to tumor progressionwas 7.6 months with 27% of patients remaining progression-free at 1 year. Themedian survival has not been reached, and the projected 1-year survival was 72%.Grade 4 neutropenia occurred in 31 patients while febrile neutropenia wasobserved in 6.

Prior Anthracycline-Based Chemotherapy

The activity of docetaxel/gemcitabine as second-linechemotherapy was also studied in patients with anthracycline-refractory and-resistant disease. Only the two Greek studies provide information concerningthe interval between first- and second-line chemotherapy to define patients with anthracycline-refractory and -resistant disease.

In the Mavroudis study,[24] anthracycline-refractory disease wasdefined as disease progression within 6 months from the beginning of front-linechemotherapy with an anthracycline-containing regimen, irrespective of theinitial response. Conversely, anthracycline-resistant disease was defined as disease progression after an initialobjective response to front-line chemotherapy with an anthracycline-basedregimen occurring more than 6 months from its initiation.

According to the inclusion criteria in the Fountzilas study,[25]all patients presented with anthracycline-resistant disease because they had tohave previous exposure to an anthracycline or mitoxantrone either in theadjuvant setting (providing they had relapsed within 12 months of completion ofchemotherapy) or for advanced disease (and subsequently progressed).

In the subgroup analysis of responses in the Mavroudis trial ofthe 32 patients who received the study treatment less than 6 months fromprevious chemotherapy, 17 responded (53%, 5 complete and 12 partial responses).Of 20 patients who received the treatment more than 6 months after their lastchemotherapy, 11 responded (55%, 2 complete and 9 partial responses).

Moreover, 1 complete and 6 partial responses (overall responserate: 46%) were obtained among 15 patients with progressive disease as bestresponse to front-line anthracycline-based chemotherapy, whereas 6 complete and15 partial responses (overall response rate: 57%) were observed among the 37patients who had an initial response before developing progressive disease.These differences were not statistically significant. In the Fountzilas trial, 3(7.5%) complete and 11 (28%) partial responses (overall response rate: 36%) wereachieved in their anthracycline-resistant patients.

Compliance With Treatment

Table 3 demonstrates the feasibility of docetaxel/gemcitabine inthe different phase II studies. The median dose intensity for docetaxel rangedfrom 21.3 to 26.5 mg/m2/wk, corresponding to a relative dose intensity rangingfrom 80% to 89%. The low relative dose intensity of docetaxel in the Mavroudisstudy should be attributed to dose reductions and/or dose delays on day 8 due togemcitabine-induced myelosuppression after its administration on day 1. Forgemcitabine, the median dose intensity in the three phase II studies ranged from486.5 to 513 mg/m2/wk, corresponding to a relative dose intensity ranging from64% to 85%.

Laufman[26] reported that 22% of gemcitabine doses were omitted,usually on day 8. This would explain the low delivered dose intensity ofgemcitabine that was observed in both the Fountzilas and Laufman studies.Conversely, in the Mavroudis trial the prophylactic administration of rhG-CSFfrom days 9 to 15 resulted in adequate bone marrow function on day 21, whichpermitted gemcitabine administration at full doses.

Toxicity

Table 4 shows the toxicity grades for these studies. Twopatients in the Mavroudis study discontinued treatment due to toxicity—onepatient had recurrent hypersensitivity reaction and one experienced prolongedgrade 3 neutropenia and thrombocytopenia. In the Fountzilas study, five patientsdiscontinued treatment—two had grade 4 neutropenia and one each had grade 3asthenia, vomiting, and diarrhea. Neutropenia was the main toxicity observed inall three trials. The incidence of grade 3 and 4 neutropenia ranged from 29% to79% of the treated patients. Moreover, febrile neutropenia occurred withouttoxic deaths in many patients enrolled in these studies. The lowest incidence ofgrade 3 and 4 neutropenia as well as febrile neutropenia observed in theMavroudis study may be attributed to the prophylactic use of rhG-CSF. Grade 4thrombocytopenia was observed in less than 5% of patients and nonhematologictoxicity was mild in all studies.

Future Directions

The data presented in this review demonstrate that thedocetaxel/gemcitabine combination is a very interesting chemotherapy regimenbased on its activity as second-line chemotherapy in patients with metastaticbreast cancer and its favorable toxicity profile. Indeed, the initial evidencefrom phase I studies evaluating different schedules of this combination showedthat the regimen is active in pretreated patients with advanced breast cancer,which was further confirmed by two multicenter and one single-institution phaseII trials. In a synthesis of the results from these three trials, the overallresponse rate was 47% associated with a median survival of more than 12 months.These efficacy results are very encouraging, because the majority (50% to 75%)of patients in these phase II studies had visceral disease and had received atleast two prior chemotherapy regimens for the treatment of metastatic disease.

In addition, all patients in the two multicenter phase IIstudies had disease refractory or resistant to anthracyclines. Moreover, almost50% of the patients enrolled in the phase II study conducted by the GreekCooperative Group for Breast Cancer (Mavroudis et al) had disease resistant orrefractory to taxanes. It is noteworthy that the docetaxel/gemcitabinecombination induced objective responses in both anthracycline- andtaxane-refractory or -resistant disease. These efficacy results were obtainedwith a mild toxicity profile. Indeed, the main adverse event observed with thedocetaxel/gemcitabine combination was grade 3 and 4 neutropenia and febrileneutropenia, which were of short duration and easily manageable with antibioticsand growth factor support. In addition, the prophylactic use of rhG-CSF in theMavroudis study decreased the incidence of both grade 3 and/or 4 neutropenia andfebrile neutropenia.

These interesting and provocative results need to be confirmed.Indeed, it is important to further define the activity of docetaxel/gemcitabinein patients with well-defined anthracycline refractory advanced breast cancer.Similarly, it seems interesting to evaluate the efficacy of the regimen inpatients with taxane (both paclitaxel [Taxol] and docetaxel)-refractory and-resistant breast cancer.

Although previous studies have shown that objective responsescould be obtained in 18% of patients with paclitaxel-resistant advanced breastcancer,[27] the observation that docetaxel/gemcitabine may induce responses inpatients who progressed under docetaxel-based front-line chemotherapy raises thequestion of an in vivo synergistic effect between the two drugs. Therefore, arandomized trial comparing the docetaxel/gemcitabine combination withgemcitabine alone in patients with docetaxel- and/or paclitaxel-refractorydisease could answer this question.

The confirmation of this encouraging activity ofdocetaxel/gemcitabine in patients with anthracycline- and taxane-refractoryadvanced breast cancer opens the way to evaluate the efficacy of this non-anthracycline-containingregimen as front-line chemotherapy in future phase II studies. Moreover, itshould be of interest to compare the activity and toxicity ofdocetaxel/gemcitabine with other chemotherapy regimens usually used asfront-line treatment of advanced breast cancer (ie,doxorubicin/cyclophosphamide, fluorouracil [5-FU]/doxorubicin/cyclophosphamide,paclitaxel/doxorubicin, docetaxel/cyclophosphamide/doxorubicin). The majoradvantage of the docetaxel/gemcitabine regimen could be the lack ofcardiotoxicity and the more predictable hematologic toxicity, which could bedecreased by the prophylactic use of rhG-CSF.

Finally, in phase I studies, some every-2-week schedules ofdocetaxel/gemcitabine administration demonstrated a favorable toxicity profileas well as activity in previously treated patients with advanced breast cancer.A dose intensification for both docetaxel and gemcitabine could be achieved withthese schedules. Although it is still unclear whether dose intensification ofthese drugs may result in an enhanced activity of the combination, theseschedules need to be further evaluated both as front-line chemotherapy and inpatients with anthracycline- and/or taxane-resistant advanced breast cancer.

References:

1. Hayes DF, Henderson IC, Shapiro CL: Treatment of metastaticbreast cancer: Present and future prospects. Semin Oncol 22(suppl 5):5-21, 1995.

2. Vogel CL: Current status of salvage chemotherapy forrefractory advanced breast cancer. Oncology 10(suppl 6):7-15, 1996.

3. Rubens RD: Routine management of disseminated disease(including anti-osteoclastic therapy) in Bonadonna G, Hortobagyi AM (eds):Textbook of Breast Cancer, p. 169. London, Gianni Murtin Dunitz, 1997.

4. Ravdin PM, Valero V: Review of docetaxel, a highly active newagent for the treatment of metastatic breast cancer. Semin Oncol 24(suppl10):18-21, 1997.

5. Ravdin PM: Docetaxel for the treatment ofanthracycline-resistant breast cancer. Semin Oncol 22(suppl 4):17-21, 1995.

6. Nabholz JM, Falkson G, Campos D, et al: A phase III trialcomparing doxorubicin and docetaxel to doxorubicin and cyclophosphamide asfirst-line chemtherapy for MBC (abstract). Proc Am Soc Clin Oncol 18:485, 1999.

7. Eisenhauer EA: Docetaxel: Current status and futureprospects. J Clin Oncol 13:2865-2868, 1995.

8. Guchelaar HJ, Richel DJ, van Knapen A: Clinical,toxicological, and pharmacological aspects of gemcitabine. Cancer Treat Rev222:15-31, 1996.

9. Merriman RL, Hertel L, Schultz RM, et al: Comparison of theantitumor activity of gemcitabine and ara-C in a panel of human breast, colon, lung and pancreatic xenograftmodels. Invest New Drugs 14:243-247, 1996.

10. Poplin EAD, Corbett T, Flaherty L, et al:Difluorodeoxycytidine (dFdC, gemcitabine): A phase I study. Invest New Drugs10:165-170, 1992.

11. Carmichael J, Walling J: Advanced breast cancer: A phase IItrial with gemcitabine. J Clin Oncol 13:2731-2736, 1995.

12. Spiridonidis CH, Laufman LR, Jones J, et al: Phase I studyof docetaxel dose escalation in combination with fixed weekly gemcitabine inpatients with advanced malignancies. J Clin Oncol 16:3866-3873, 1998.

13. Kaye SB: Gemcitabine: Current status of phase I and IItrials. J Clin Oncol 12:1527-1531, 1994.

14. Rischin D, Boyer M, Smith J, et al: A phase I trial ofdocetaxel and gemcitabine in patients with advanced cancer. Ann Oncol11:421-426, 2000.

15. Eckardt JR, Schmidt AM, Needles BM, et al: A phase I studyof the combination of docetaxel and gemcitabine (abstract). Proc Am Soc ClinOncol 17:920, 1998.

16. Rigas JR, Rothenberg ML, Davis TH, et al: Phase I clinicaland pharmacokinetic study of docetaxel in combination with two gemcitabineinfusion schedules (abstract). Proc Am Soc Clin Oncol 18:870, 1999.

17. Zoli W, Ricotti L, Dal Susino M, et al: Docetaxel andgemcitabine activity in NSCLC cell lines and in primary cultures from human lungcancer. Br J Cancer 81:609-615, 1999.

18. Denes AE, Needles BM, Schmidt A, et al: A comparison of twoschedules of docetaxel in combination with gemcitabine given every other week(abstract). Proc Am Soc Clin Oncol 19:828, 2000.

19. Burstein HJ, Manola J, Younger J, et al: Docetaxeladministrered on a weekly basis for metastatic breast cancer. J Clin Oncol18:1212-1219, 2000.

20. Kouroussis Ch, Agelaki S, Mavroudis D, et al: Adose-escalating study of weekly docetaxel in patients with advanced solidtumors. Cancer Chemother Pharmacol. In press.

21. Brugnatelli S, Danova M, Tamburo M, et al: Weekly scheduleof gemcitabine plus docetaxel in refractory advanced breast cancer: Adose-finding study (abstract). Proc Am Soc Clin Oncol 19:6091, 2000.

22. Kouroussis C, Kakolyris S, Stathopoulos G, et al: A phase Istudy of weekly docetaxel and gemcitabine in advanced non-small cell lungcancer. Proceedings of the 9th Conference on Lung Cancer, Tokyo, September 9-15,2000. Lung Cancer 29(suppl 1):43, 2000.

23. Frasci G, Comella P, D’Aiuto G, et al: Weekly docetaxelplus gemcitabine or vinorelbine in refractory advanced breast cancer patients: Aparallel dose-finding study. Ann Oncol 11:367-371, 2000.

24. Mavroudis D, Malamos N, Alexopoulos A, et al: Salvagechemotherapy in anthracycline-preteated patients with docetaxel and gemcitabine: A multicenter phase II trial. Ann Oncol 10:211-215, 1999.

25. Fountzilas G, Nicolaides C, Bafaloukos D, et al: Docetaxeland gemcitabine in anthracycline-resistant advanced breast cancer: A HellenicCooperative Oncology Group phase II study. Cancer Invest 18:503-509, 2000.

26. Laufman LR, Spiridonidis CH, Carman L, et al: Second-linechemotherapy with weekly gemcitabine and monthly docetaxel in patients withmetastatic breast cancer (abstract). Proc Am Soc Clin Oncol 19:408, 2000.

27. Valero V, Jones SE, Von Hoff DD, et al: A phase II study ofdocetaxel in patients with paclitaxel-resistant metastatic breast cancer. J ClinOncol 16:3362-3368, 1998.