Adoptive Immunotherapy For Superficial Bladder Cancer With Autologous Macrophage Activated Killer Cells

0022-5347/02/1686-2373/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 168, 2373–2376, December 2002 Printed in U.S.A.

DOI: 10.1097/01.ju.0000038342.03540.0a

Original Articles ADOPTIVE IMMUNOTHERAPY FOR SUPERFICIAL BLADDER CANCER WITH AUTOLOGOUS MACROPHAGE ACTIVATED KILLER CELLS NICOLAS THIOUNN, FRANCK PAGES, ARNAUD MEJEAN, JEAN-LUC DESCOTES, WOLF-H. FRIDMAN AND JEAN-LOUP ROMET-LEMONNE* From the Service d’Urologie, Hoˆpital Cochin, Service d’Immunologie Biologique, Hoˆpital Europe´en Georges Pompidou, Service d’Urologie, Hoˆpital Necker, Service de Chirurgie Urologique, Hoˆpital Michallon, Grenoble, and Immuno-Designed Molecules, Paris, France

ABSTRACT

Purpose: We assessed the efficacy and safety of adoptive immunotherapy administered to 17 patients with TaGIII or recurrent TaGII superficial bladder cancer following transurethral tumor resection. Materials and Methods: Macrophage activated killer (MAK) cells were obtained from autologous mononuclear cells harvested by apheresis, after in vitro culture for 7 days and activation with interferon-␥ on the last day of culture. The patients received 6 weekly intravesical infusions of approximately 2 ⫻ 108 cells each. Additionally, 5 patients received 2 or 3 more infusions at 3-month intervals. Each patient was followed for 1 year or until tumor recurrence, whichever came first. Results: A total of 112 intravesical infusions were performed. During the 12-month followup period 8 patients experienced 11 common toxicity criteria grade 1 or grade 2 adverse events considered possibly related to protocol. No clinically relevant grade 1 or 2 laboratory test results were reported while the patients received treatment. In 17 patients 8 tumors recurred compared to 34 recurrences during the year before the first MAK cell infusion. This difference was highly significant (p ⱕ0.0005). Conclusions: The promising efficacy and safety results of this study and the fact that the MAK cell treatment regimen proved feasible should encourage initiation of further large scale studies to confirm these data. KEY WORDS: immunotherapy, bladder neoplasms; administration, intravesical; macrophage activation

Activated macrophages are capable of selectively lysing tumor cells, in particular after activation with interferon␥.1, 2 The antitumoral properties of activated macrophages, or macrophage activated killer (MAK) cells, have been demonstrated in vitro in murine experimental models including nude mice xenografted with human tumors.3, 4 MAK cells can be obtained by in vitro culture of blood monocytes harvested by apheresis. Standardized procedures and quality controlled devices (MAK Cell Processor, IDM, Paris, France) have been developed to make the complete process reproducible and to control the quality of the final cell preparations.5 Clinical trials conducted on a variety of patients with cancer with systemic (intravenous) or local (intraperitoneal or intrapleural) administrations of macrophages have shown that these preparations are well tolerated. Although clinical evidence of an antitumor effect of macrophage adoptive therapy is rather limited, a few partial responses and disease stabilizations have been reported in various types of advanced solid tumor cancer.5, 6 The in vitro cytotoxicity mediated by macrophages appears to require direct cell-to-cell interaction resulting in membrane disruption and lysis of target tumor cells, although there is some evidence that soluble macrophage derived products may also lyse some tumor cells.7 However, experiments in murine models of solid tumors show that intratumoral injection of macrophages gives better results than systemic injection.8 This finding suggests that MAK cell therapy might be more effective when administered locally. A conventional treatment of superficial bladder tumors Accepted for publication June 21, 2002. * Financial interest and/or other relationship with IDM.

includes transurethral resection followed by intravesical infusion of a bacillus Calmette-Guerin (BCG) preparation as preventive therapy for recurrence or progression. Recurrences most often develop within 12 months although they may be diagnosed several years after transurethral resection. Increased risk of recurrence has been associated with large tumor size, advanced stage, higher grade, multifocal and recurrent character of the tumor or presence of carcinoma in situ.9, 10 Several studies have shown the therapeutic benefit provided by BCG compared to transurethral resection alone,11, 12 although 20% to 30% of recurrences are still observed after BCG therapy combined with resection. The mechanism by which BCG prevents tumor recurrence is not clearly established. Evidence suggests that interferon-␥ activation of macrophages is implicated.13 Intravesical infusions of BCG result in significant local morbidity, in particular vesicourethral irritation, but also fever and flu-like symptoms, hematuria, acute bacterial prostatitis or BCG induced granulomatous prostatitis, epididymitis, lower urethral obstruction and pseudo-tuberculous small bladder.14 In view of the few and mild expected side effects due to MAK cell therapy and the possibility to inject MAK cells at the tumor site, we conducted a study of intravesical infusion of ex vivo interferon-␥ activated autologous MAK cells in patients with superficial bladder tumors. The main objectives of this phase I/II study were to determine the feasibility, toxicity and antitumoral activity of this adoptive immunotherapy. This study was conducted in patients with TaGIII or relapsed TaGII superficial bladder cancer, as these stages are often associated with relapse within less than 1 year.9, 10

2373

2374

MAK CELLS FOR TREATMENT OF BLADDER CANCER MATERIALS AND METHODS

Patient population. A total of 17 patients were treated with MAK cells. Shortly before treatment they had undergone transurethral resection for unifocal or plurifocal superficial TaGIII or recurrent TaGII vesical tumors with no associated carcinoma in situ as documented on anatomic pathological examination. Eligibility requirements also included a WHO performance status of 2 or less, blood creatinine 200 ␮mol./L. or less, normal blood transaminases and white blood cell count 3,500/mm.3 or greater. Exclusion criteria were serious concomitant conditions, such as cardiac or respiratory failure, severe infection, as well as pregnancy or lack of effective contraception in female patients with childbearing potential, positive HIV, hepatitis B virus or hepatitis C virus serology, psychiatric disorders, autoimmune disease or immunodeficiency, history of malignant tumor and evidence of tumor recurrence at the fibroscopic examination scheduled immediately before initiation of MAK cell treatment. The study was approved by a local ethics committee and signed informed consent was obtained from all patients. Study design. Apheresis was scheduled within 1 month after complete resection of the vesical tumor and anatomic pathological characterization. Harvested mononuclear cells were cultured for 6 days for differentiation into macrophages. On day 6 of cell culture macrophages were activated into MAK cells using interferon-␥. The cells were aliquoted into 6 bags each of 3 ⫻ 108. Of the 6 bags 5 were frozen and 1 was immediately infused intravesically to the corresponding patient. The 5 other preparations were infused after thawing at weekly intervals. After a protocol amendment to perform maintenance therapy, 5 patients received 2 or 3 additional intravesical infusions at months 3, 6 and 9, respectively. For these patients activated MAK cells were aliquoted into 9 bags. If the initial total MAK cell count was insufficient, a second apheresis was performed to prepare the additional bags. Patients were followed for 12 months following the first infusion. Clinical examinations and laboratory tests were performed at study inclusion and before each MAK cell infusion. Cystoscopy and urinary cytology were done at 3, 6, 9 and 12 months. Toxicity was monitored throughout the 12month followup. MAK cell treatment. For the preparation of MAK cells, approximately 10 ⫻ 109 mononuclear cell enriched leukocytes were first obtained by apheresis. Using a MAK Cell Processor, they were cultured in macrophage medium medium supplemented with 3% to 5% of uncomplemented autologous serum for 6 days. Macrophages were activated with interferon-␥. MAK cells were elutriated and identified by microscopic examination of morphology and by assessment of CD14, CD64 and HLA-DR antigen expression. Bacteriological controls were performed throughout the process and immediately before infusion of each MAK cell bag. MAK cell preparations were infused intravesically at a volume of 50 ml. after emptying the bladder. Patients were encouraged to keep the MAK cell preparation in the bladder for 2 hours. Safety and efficacy evaluation. All adverse events spontaneously reported or observed were recorded. Severity was recorded using common toxicity criteria and the relationship to study protocol was assessed by the investigator. Toxic effects as determined by laboratory tests (complete blood count, platelet count, blood electrolytes, creatinine, blood coagulation, transaminases and bilirubin) were also graded according to common toxicity criteria. Antitumoral activity of MAK cell treatment was assessed by cystoscopy. The number of recurrences during the year after beginning treatment was recorded. Statistical analyses. Efficacy of the MAK cell treatment was assessed by comparison of recurrence rates during the year after and the year before the first infusion, respectively. We performed 3 different statistical tests. In the first 2 tests adjusted mean resection rate (r⬘) during the previous year was

estimated as a fraction of the number of resections per patientyear at risk. For this fraction the denominator was the sum of the individual differences in years between the first infusion date and the earliest event date during the previous year for all 17 patients, and the numerator was the total number of resections after the earliest one during that year. The probability to observe at least 1 recurrence during the previous year (given the mean resection rate during the previous year) was expressed by Poisson’s distribution (p ⫽ 1 ⫺ er⬘). The percentage of patients with recurrence during the 1-year study was compared to the expected p value of recurrences using an exact binomial test. The individual resection frequency for the year after the beginning of treatment was estimated from {[the number of resections after the earliest one during the preceding year] ⫹ [(time remaining in the year)*r⬘]}. The difference between this estimated resection frequency for a 1-year period before MAK cell treatment and the observed number of recurrences during the study year was compared to 0 using the Wilcoxon signed rank test. In a third statistical approach the difference between the numbers of events observed during the previous year and the study year was evaluated with the Wilcoxon signed rank test. RESULTS

Patient characteristics. The characteristics of the 16 males and 1 female are summarized in table 1. Most patients had TaGII tumors as documented at transurethral resection before the first MAK cell infusion, and 4 had TaGII disease during the previous year (primary tumor occurrence in 2). The number of tumor sites before study ranged from 1 to 4. Of the 17 treated patients 6 had received previous BCG therapy. Cystoscopy before the first MAK cell infusion on day 7 was normal in 11 of the 17 patients. Of the remaining 6 patients 1 had abnormal results, a small lesion, 2 each had each 3 polyps that were electrocoagulated and 3 had scar at the last transurethral resection site without any sign of recurrence. Exposure to MAK cells. Exposure of the bladder wall to MAK cells is summarized in table 2. The 17 patients received a total of 112 infusions. For the 10 patients who received 6 infusions, mean MAK count per infusion ranged from 14.2 ⫻ 107 to 25.5 ⫻ 107 for a total exposure time of 11.4 to 13.8 hours. For the 4 patients who received 9 infusions mean MAK count per infusion ranged from 9.8 ⫻ 107 to 21.7 ⫻ 107 for a total exposure time of 17.4 to 18.2 hours. Treatment was discontinued in 3 patients because of tumor recurrence after 3, 5 and 8 infusions, respectively. Safety results. A total of 13 adverse events were experienced by 9 patients (52.9%) during the 12-month followup, (table 3) the most frequent of which were dysuria, hematuria and urinary tract infection (table 3). Prostatic disorder was reported in 2 patients. Common toxicity criteria results did not exceed grade 2 except for a grade 3 vesical neoplasm which required hospitalization. All but 2 adverse events (an accidental fall and the vesical neoplasm) were considered possibly related to the study protocol by the investigator. TABLE 1. Patient characteristics No. pts. No. male/female Median age (range) No. of occurrences previous yr. (No. pts.): TaGII* TaGI/II TaGIII TaGII/III No. tumors before study (No. pts.): TaGII TaGI/II TaGIII TaGII/III * First occurrence for 2 patients.

17 16/1 65 (53–75) 1, 2, 3, 4 3 1 1

(4, 3, 2, 2) (2) (3) (1)

1, 3, 4, Not determined (4, 1, 3, 3) 3, 4 (1, 1) 2, Not determined (2, 1) 3 (1)

MAK CELLS FOR TREATMENT OF BLADDER CANCER TABLE 2. Extent of exposure of the bladder wall to MAK No. Infusions

No. Pts.

Mean MAK Count/Infusion (107)

3 5 6 8 9

1 1 10 1 4

9.2 12.3 14.2–25.5 16.3 9.8–21.7

Total Hrs. Bladder Exposure to MAK Cells 5.4 10.3 11.4–13.8 16.0 17.4–18.2

TABLE 3. Adverse events reported under MAK cell treatment during 12-month followup Adverse Event

2375

rate compared to the expected rate (0.47 versus 0.87, p ⫽ 0.0002). When comparing the estimated individual resection rates per year to the observed number of recurrences during the study year using a signed rank test, the differences between the 2 values were significantly different from 0 (p ⬍0.0001) in favor of a lower actual recurrence rate compared to the estimated resection rate. In addition, comparison of the resection rate in the 17 patients during the year before the first MAK cell infusion to that during the year after the first infusion using a signed rank test also revealed a significantly lower rate of events with treatment (p ⫽ 0.0005).

Common Toxicity Criteria 1

2

Myalgia 1 Dysuria 2 Hematuria 1 Urinary tract infection 2 1 Prostatic disorder 1 1 Neoplasm Headache 1 Fatigue 1 Fall 1* * Adverse events not considered related to study protocol.

3

1*

Study protocol related adverse events were reported in 8 patients. The laboratory parameters were not significantly modified by MAK cell treatment. Most abnormal values from treatment did not exceed common toxicity criteria grade 1. Only 1 grade 3 lymphopenia was observed after MAK cell treatment in a patient with grade 2 common toxicity criteria at days 0 and 7. Thus, the changes from baseline in laboratory common toxicity criteria grades with MAK cell treatment were minimal and probably without clinical relevance. Efficacy results. The occurrence of vesical tumors in the patient during the year before and the year after the first MAK cell infusion is illustrated in the figure. Of the 17 patients 8 had recurrence during the 1-year followup after the first infusion for a tumor recurrence rate of 0.47 with a 95% CI of 0.23, 0.72. Results of comparisons of vesical tumor recurrence during the year before and the year after the first infusion are summarized in table 4. Comparison of the percentage of patients with 1 tumor recurrence during the study year to the expected rate of recurrence, using an exact binomial test, showed a statistically significant lower incidence

Occurrence of vesical tumors by patient during year before and year after first MAK cell infusion and during subsequent followup. Asterisks indicate maintenance therapy consisting of 2 additional (COCH18) or 3 additional infusions at 3-month intervals.

DISCUSSION

The high potential for recurrence or progression of superficial bladder tumors has led to widespread use of adjuvant intravesical instillations of BCG. The mechanisms by which BCG mediates antitumor effect are not fully understood but the implication of a local immune response is attested by a marked infiltration of CD4⫹ lymphocytes and macrophages in the bladder wall.15 Whereas tumor specific immunity after BCG therapy remains controversial, BCG induces soluble factors that might favor elimination of tumors by macrophage mediated cytotoxic mechanisms, for example by interferon-␥ which can be detected in large quantities in urine following treatment.16 Interleukin 8 and -18 cytokines secreted by activated macrophages were shown to correlate with a good prognosis when detected in urine 12 hours after BCG therapy.17 These findings supporting a significant role for macrophages in the BCG mediated immune response prompted us to initiate a clinical evaluation of interferon-␥ activated autologous macrophages. An objective of this pilot study was to establish the feasibility of this novel treatment by MAK cells. The planned time requirements between transurethral resection and first injection could not be strictly met and most patients had the first infusion 48 days or more after the last tumor resection. This delayed treatment initiation might have negatively influenced the efficacy results by postponing a potentially effective treatment. Otherwise, only 1 patient had a minor deviation in the infusion schedule. The MAK cell treatment regimen, in terms of infusion scheduling and volume infused, was similar to that reported for BCG therapy in patients with superficial bladder cancer.18, 19 Unlike BCG treatment after transurethral resection, which requires a previous careful evaluation of the risk of serious systemic complications and may result in treatment discontinuation or intensive care when such complications do occur,14 the MAK cell preparation could be administered to all patients in the study and was well tolerated. None of the treated patients withdrew from the study due to the treatment procedure or for other tolerability reasons. Only 8 of the 17 patients had 1 tumor recurrence during the 1-year followup. Using tumor recurrence frequency during the year before the first MAK cell infusion for comparison was based on the hypothesis that, in the absence of adjunctive treatment to transurethral resection, the probability for tumor recurrence would remain unaltered. The estimated tumor recurrence rate in the study population during the previous year (uncorrected 2.0 or adjusted 2.05 recurrences per year) was greater than the yearly recurrence rate of 1.7 reported in the literature for similar tumors in the absence of BCG treatment.20 Regardless of the statistical approach, the difference between pretreatment and followup years was highly significant in favor of the efficacy of adjuvant MAK cell treatment. However, the inclusion of 2 patients after a first occurrence of a TaGII tumor may have biased the efficacy results, since these patients did not have any recurrence during the 12 months after the first infusion and even during the subsequent followup. Although the efficacy analyses were performed in a small population of patients and the aforementioned protocol deviations may have introduced a bias, the

2376

MAK CELLS FOR TREATMENT OF BLADDER CANCER TABLE 4. Comparison of tumor resection rate between the year before and the year after first MAK cell infusion Yr. Before First Infusion

Yr. After First Infusion

No. of Events (A)

No. of Events After First Event (B)

Time Between First Event And First Infusion (yrs.)

Time Remaining In Yr. (yrs.) (C)

Estimated Individual Rate Per yr. (D)

No. of Recurrences (E)

A–E

D–E

NECK02 CHRG01 COCH01 COCH03 COCH04 COCH05 COCH07 COCH08 COCH09 COCH10 COCH11 COCH12 COCH14 COCH15 COCH16 COCH17 COCH18

1 3 4 2 4 2 1 1 1 1 1 3 3 1 3 1 2

0 2 3 1 3 1 0 0 0 0 0 2 2 0 2 0 1

0.55 0.87 0.82 0.48 0.85 0.61 0.16 0.36 0.16 0.17 0.13 0.83 0.96 0.11 0.78 0.18 0.31

0.45 0.13 0.18 0.52 0.15 0.39 0.84 0.64 0.84 0.83 0.87 0.17 0.04 0.89 0.22 0.82 0.69

0.93 2.27 3.37 2.07 3.31 1.80 1.73 1.31 1.73 1.71 1.77 2.35 2.09 1.83 2.44 1.69 2.42

1 1 1 0 1 0 1 1 0 0 1 0 0 0 0 1 0

0 2 3 2 3 2 0 0 1 1 0 3 3 1 3 0 2

⫺0.07 1.27 2.37 2.07 2.31 1.80 0.73 0.31 1.73 1.71 0.77 2.35 2.09 1.83 2.44 0.69 2.42

Totals

34

17

8.31

8.67

34.82

8

26

26.82

Pt. No.

Statistical r ⫽ 2* r⬘ ⫽ 2.05† (0.47 [0.23, 0.72]),‡ p ⫽ 0.0005 p ⬍0.0001 results p ⫽ 0.0002 (signed rank test) (signed rank test) * Total of 34 resections per 17 patient-years. † Total of 17 resections per 8.31 patient-years, and probability of having at least 1 event in the followup year assuming same patient conditions as for the first event in the preceding year is p ⫽ 1 ⫺ e 2.05 ⫽ 0.87 [Poisson’s distribution]. ‡ 95% CI and 80% CI was 0.30, 0.65.

comparison of the tumor recurrence rates before and after the first MAK cell infusion in a year suggests that this treatment may be effective against recurrence of superficial TaGII/III bladder tumors in our patients. CONCLUSIONS

In this phase I/II study no severe protocol related adverse events or discontinuations due to adverse events were reported among 17 patients treated with 6 weekly intravesical MAK cell infusions plus 2 or 3 additional monthly infusions in 5. Furthermore, safety was satisfactory, with no systemic complications. The efficacy results indicate that MAK cell therapy may decrease the number of recurrences in patients with superficial TaGII/III bladder tumor in the 12 months after initiation of treatment. These preliminary promising results and the fact that the MAK cell treatment regimen proved feasible should encourage initiation of further large scale studies to confirm these safety and efficacy data. REFERENCES

1. Chokri, M., Lopez, M., Oleron, C., Girard, A., Martinache, C., Canepa, S. et al: Production of human macrophages with potent antitumor properties (MAK) by culture of monocytes in the presence of GM-CSF and 1,25-dihydroxy (OH)2 vitamin D3. Anticancer Res, 12: 2257, 1992 2. Gordon, S., Clarke, S., Greaves, D. and Doyle, A.: Molecular immunobiology of macrophages: recent progress. Curr Opin Immunol, 7: 24, 1995 3. Bartholeyns, J., Lombard, Y., Dumont, S., Hartman, D., Chokri, M., Giaimis, J. et al: Immunotherapy of cancer: experimental approach with activated macrophages proliferating in culture. Cancer Detect Prev, 12: 413, 1988 4. Fidler, I. and Kleinerman, E. S.: Therapy of cancer metastasis by systemic activation of macrophages: from the bench to the clinic. Res Immunol, 144: 284, 1993 5. Bartholeyns, J., Romet-Lemonne, J. L., Chokri, M. and Lopez, M.: Immune therapy with macrophages: present status and critical requirements for implementation. Immunobiology, 195: 550, 1996 6. Lesimple, T., Moisan, A., Guille´ , F., Leberre, C., Audran, R., Drenou, B. et al: Treatment of metastatic renal cell carcinoma with activated autologous macrophages and granulocyte—macrophage colony-stimulating factor. J Immunother, 23: 675, 2000 7. Lesimple, T., Moisan, A. and Toujas, L.: Autologous human macrophages and anti-tumour cell therapy. Res Immunol,

149: 663, 1998 8. Chokri, M., Freudenberg, M., Galanos, C., Poindron, P. and Bartholeyns, J.: Antitumoral effects of lipopolysaccharides, tumor necrosis factor, interferon and activated macrophages: synergism and tissue distribution. Anticancer Res, 9: 1185, 1989 9. Heney, N. M., Ahmed, S., Flanagan, M. J., Frable, W., Corder, M. P., Haferman, M. D. et al: Superficial bladder cancer: progression and recurrence. J Urol, 130: 1083, 1983 10. Prout, G. R., Jr., Barton, B. A., Griffin, P. P. and Friedall, G. H. for the National Bladder Cancer Group: Treated history of non invasive grade 1 transitional cell carcinoma. J Urol, 148: 1413, 1992 11. Morales, A., Eidinger, D. and Bruce, A. W.: Intracavitary bacillus Calmette-Guerin in the treatment of superficial bladder tumors. J Urol, 116: 180, 1976 12. Lamm, D. L.: Long-term results of intravesical therapy for superficial bladder cancer. Urol Clin North Am, 19: 573, 1992 13. Patard, J. J., Muscatelli-Groux, B., Saint, F., Popov, Z., Maille, P., Abbou, C. et al: Evaluation of local immune response after intravesical bacillus Calmette-Guerin treatment for superficial bladder cancer. Br J Urol, 78: 709, 1996 14. Lamm, D. L.: Complications of bacillus Calmette-Guerin immunotherapy. Urol Clin North Am, 19: 565, 1992 15. Bohle, A., Gerdes, J., Ulmer, A. J., Hofstetter, A. G. and Flad, H.-D.: Effects of local bacillus Calmette-Guerin therapy in patients with bladder carcinoma on immunocompetent cells of the bladder wall. J Urol, 144: 53, 1990 16. Prescott, S., James, K., Hargreave, T. B., Chisholm, G. D. and Smyth, J. F.: Radio-immunoassay detection of interferongamma in urine after intravesical Evans BCG therapy. J Urol, 144: 1248, 1990 17. Thalmann, G. N., Sermier, A., Rentsch, C., Mohrle, K., Cecchini, M. G. and Studer, U. E.: Urinary interleukin-8 and 18 predict the response of superficial bladder cancer to intravesical therapy with bacillus Calmette-Guerin. J Urol, 164: 2129, 2000 18. Mulders, P. F., Meyden, A. P., Doesburg, W. H., Oosterhof, G. O. and Debruyne, F. M.: Prognostic factors in pTa-pT1 superficial bladder tumours treated with intravesical instillations. The Dutch South-Eastern Urological Collaborative Group. Br J Urol, 73: 403, 1994 19. Eure, G. R., Cundiff, M. R. and Schellhammer, P. F.: Bacillus Calmette-Guerin therapy for high risk stage T1 superficial bladder cancer. J Urol, 147: 376, 1992 20. Kamat, M. R., Kulkarni, J. N., Tongaonkar, H. B. and Dalal, A. V.: Intravesical bacillus Calmette-Guerin for superficial bladder cancer: experience with Danish 1331 strain. J Urol, 152: 1424, 1994