Short-term tumor cell lines from breast cancer for use as autologous tumor cell vaccines in the treatment of breast cancer.

OBJECTIVE: We tried to establish short-term cultures of autologous tumors from patients with breast carcinoma for potential use as active specific immunotherapy (i.e., autologous vaccine) after resection of primary breast cancer, and/or for the treatment of metastases. METHODS: Between 10/90 and 12/99 the cell biology laboratory of the Hoag Cancer Center attempted to establish short-term tumor cell lines from 115 breast cancer specimens from 56 primary breast lesions, 17 axillary nodes, 14 other lymph node/soft tissue sites, 10 chest wall recurrences, and 6 thoracentesis of malignant pleural effusions. Success was defined by growth of 5 x 10(7) viable cells whose malignant nature and breast cancer origin was confirmed by histology of the submitted tissue, cell morphology and antigenic phenotyping. Variables associated with successful growth of short-term cell lines were examined. RESULTS: Expansion to 5 x 10(7) cells was achieved for only 8/115 samples [7%] including two from chest wall recurrences, and one each from a supraclavicular node, an umbilical node, liver, omentum, and pleural fluid. Two of the successful cell lines were established from tissue that originally had been cryopreserved; the others were initiated from fresh tumor. The success rate was better from regional/distant metastases 7/55 (13%) compared to primary tumors 1/56 (1.8%) (p = 0.063). The success rate for tumors harvested at Hoag Hospital was 4/97 (4%) compared to 4/14 from (31%) distant sites, but all but one of the tumors from a distant geographic site was a metastatic lesion. Tumor cell lines were successfully established from metastatic lesions ranging in size from < 1.0 g to 19 g. Four patients were treated with their autologous vaccine in the setting of chemotherapy-refractory metastatic disease without any significant toxicity. CONCLUSIONS: We were unable to establish short-term cell lines for most patients with primary or metastatic breast cancer using this methodology. However, two long-term cell lines have been established and characterized. Treatment with the autologous irradiated cell product was not associated with acute toxicity.

Short-term autologous tumor cell lines for the active specific immunotherapy of patients with metastatic melanoma.

We established short-term cell lines for 108/170 (64%) patients with metastatic melanoma. Tumor cell numbers were expanded to 10(8), then cells were irradiated, aliquoted, and cryopreserved for clinical use. Vaccines have been used to treat 69 patients with clinical follow up for 33 who had measurable metastatic disease at the time vaccine therapy was initiated (METS), and 33 who had no evidence of disease (NED) at the time of vaccine therapy following surgical resection of metastases. The protocol called for a baseline test of delayed tumor hypersensitivity (DTH), three weekly injections, a repeat of the DTH test, then monthly injections for an additional 5 months. Objective tumor responses were noted in 3/26 (12%) patients who received a minimum of three vaccinations, one complete, and two partial, with survivals of 36, 46+, and 78+ months. Only 6/64 (9.4%) had a positive DTH (>10 mm) at baseline, including three METS, all of whom progressed within 4 months and died within a year, and three who are still NED after more than 5 years. Conversion of DTH from negative to positive was documented in 18/44 (41%) patients who were tested at week 0 and 4. At a median follow up of greater than 5 years, the median overall survival (OS) was 40 months for "NED" with a 5-year survival rate of 39%, and 8.6 months with a 5-year survival rate of 10% for "METS" The 18 patients who had conversion of their DTH had a median event-free survival (EFS) of 15.8 months and 5-year EFS of 32% compared to 4.2 months and 9% for the 26 non-converters (P=0.012, two-tailed, log-rank test). Among patients who were NED when treatment started, the 12 patients whose DTH converted had a median overall survival of 61.4 months with 5-year survival of 63% compared to 9.7 months and 0% for the 13 non-converters (P=0.0026). This treatment approach is feasible, produces minimal toxicity, and is associated with long-term survival in a significant subset of patients.

Characterization of human tumor-infiltrating lymphocytes expanded in hollow-fiber bioreactors for immunotherapy of cancer.

We attempted to grow tumor-infiltrating lymphocytes (TIL) from 34 fresh tumors of eight different histologies using flasks for the initiation phase and hollow fiber bioreactors to expand TIL to therapeutic numbers. Overall success rate was 76% (26/34) including melanoma (9/14, 64%) and renal cell carcinoma (11/11, 100%). The mean number of days required to reach successful initiation (1 x 10(9) TIL) for all tumor types was 29 +/- 16 days (mean +/- S.D.). Therapeutic doses of TIL required an average of 88 +/- 23 days (initiation plus expansion) with an average TIL number of 3.2 x 10(10) +/- 2.8 x 10(10). TIL phenotype was predominantly CD4+ in 53% (16/30) and CD8+ in 47% (14/30), renal cell carcinoma samples accounted for 12/14 of the predominantly CD8+ TIL. Cells bearing the natural killer (NK) phenotype represented only 0-7% of TIL while LAK phenotype represented 0-68% (mean 11 +/- 15%); LAK was the predominant phenotype in one patient with kidney cancer. Cytotoxicity tests showed consistent NK and LAK activity in addition to cytolysis of autologous tumor. Autologous tumor cell restricted cytolysis was noted for three TIL cultures. The overall success rate and characteristics of TIL were similar to our results with TIL expanded in semi-permeable plastic bags. This work confirms that hollow-fiber bioreactors are a suitable alternative to semi-permeable bags and roller bottle systems for the expansion of human TIL for therapeutic use in cancer patients.

Short-term tumor cell lines from renal cell carcinoma for use as autologous tumor cell vaccines in the treatment of kidney cancer.

OBJECTIVE: We have tried to establish short-term cultures of autologous tumors from patients with renal cell carcinoma that could be used as active specific immunotherapy (i.e., autologous vaccine) in such patients after resection of primary kidney cancer, and/or for the treatment of metastatic cancer. METHODS: Between 10/90 and 9/99 the cell biology laboratory of the Hoag Cancer Center received 69 kidney tumor samples that had been surgically excised, including 43 primary tumors and 26 metastatic lesions. Efforts were made to establish short-term tumor cell cultures, as defined by the growth of 10(8) cells; malignant nature and renal cell origin were confirmed by morphology and antigenic phenotyping. Variables associated with successful growth of short-term cell lines were examined. RESULTS: Short-term cell lines were successfully established from 55/69 samples [80%] including 36/43 (84%) from primary tumors and 19/26 (73%) from metastatic lesions. The success rate for tumors harvested at Hoag Hospital was 40/50 (80%); the success rate for tumors obtained from other geographic areas was 15/19 (79%). Tumor cell lines were successfully established from metastatic lesions ranging in size from a 0.5 g vertebral lesion to a 22 g rib/lung chest wall metastasis, and from primary renal cell lesions ranging in size from 1.5 g to 39.7 g. CONCLUSIONS: Short-term cell lines can be established for most patients with primary or metastatic renal cell carcinoma making a pure autologous tumor-cell vaccine approach feasible. Vaccines have been prepared for 41 patients and a vaccine therapy trial is in progress.

The feasibility of using short-term cultures of ovarian cancer cells for use as autologous tumor cell vaccines as adjuvant treatment of advanced ovarian cancer.

OBJECTIVE: We have tried to establish short-term cultures of autologous tumors from patients with stage III and IV ovarian cancer, which could be used as active specific immunotherapy, (i.e., autologous vaccine) in such patients after debulking surgery & combination chemotherapy. METHODS: Between 5/93 and 11/97 the Hoag cell biology laboratory received 53 ovarian tumor samples that had been surgically excised at the time of laparotomy, and four samples of malignant ascites. Efforts were made to establish short-term tumor cell cultures as confirmed by morphology & phenotype. RESULTS: Short-term proliferating cultures were successfully established from 21/57 samples [37%] which included 8/24 [33%] successes from samples obtained at diagnosis compared to 13/33 [37%] samples obtained at the time of a relapse [p = 0.45]. The probability of successful culture was not related to tumor size for samples with a range of 0.8-34 g (mean 5.8 g). One patient was treated in the setting of metastatic disease and one in the adjuvant setting; both received repeated injections of irradiated autologous tumor cells plus granulocyte macrophage stimulating factor (GM-CSF). In one patient a delayed tumor hypersensitivity skin test converted from negative to positive. CONCLUSIONS: Short-term cultures of autologous tumor cells for use as tumor cell vaccines can be established for about one-third of patients with ovarian cancer using this methodology and the treatment approach is feasible.

Clinical experience with autologous tumor cell lines for patient-specific vaccine therapy in metastatic melanoma.

Because of their patient specificity and proliferative capacity, tumor cell lines established from autologous metastatic melanoma tumor samples may be an excellent immunogen for patient-specific vaccine therapy. Between October 1990 and July 1996, the Hoag Cancer Center cell biology laboratory received 136 fresh metastatic melanoma samples from 122 different patients. Tumor cell lines were successfully established for 92 of 136 samples (68%), for 87 of 122 patients (71%). Successful cultures were expanded to 10(8) cells (total culture time about 8 weeks), confirmed to be sterile, irradiated, and stored frozen in aliquots of 10(7) cells. Vaccines were prepared from 72 lines, and 62 vaccines were used in 57 different patients. Subcutaneous vaccination took place on weeks 1, 2 and 3, and then monthly for a total of 6 months. A delayed tumor hypersensitivity skin test (DTH) was administered at week zero and week 4. Various adjuvants were co-administered including BCG, alpha- or gamma-interferon, and GM-CSF. Patients were monitored for failure-free survival (FFS) and overall survival (OS) from the date of the first vaccination. Follow-up data is available for 52 patients, 27 who had no evident disease (NED) at the time of vaccination and 25 who had metastatic disease at the time of treatment. There were two partial responses which persisted 11.9 and 39.8+ months among the 25 patients who had detectable metastatic disease whün treatment was initiated (8%, 1 to 26%, 95%-Ci). Twenty patients had negative skin tests at week 0 and week 4; six were positive both times, and 13 converted their DTH from negative to positive, for a conversion rate of 13 of 33 (39%). Patients who received interferon-gamma and/or GM-CSF as an adjuvant had a higher rate of DTH conversion compared to patients who received other adjuvants (13 of 20 v 2 of 13, P = 0.003). For patients who were NED, nine of 19 (47%) converted their DTH test compared to four of 14 (29%) patients with metastatic disease (p = 0.33). For patients whose DTH converted from negative to positive after 3 weeks of vaccination, median FFS and OS were superior compared to patients whose DTH remained negative (19.4 v 4.0 months FFS, p = 0.0052 and 39.6 v 18.3 months OS, p = 0.0602). The autologous cell line approach to active specific immunotherapy is feasible for patients who have resectable foci of metastatic disease. Administration of such patient-specific vaccines improves survival for those patients who are NED at the time of vaccination and convert their DTH skin test, compared to those whose DTH test remains negative.

Characterization of tumor-infiltrating lymphocytes derived from human tumors for use as adoptive immunotherapy of cancer.

From 1991 to 1995, we initiated cultures of 94 fresh tumor samples of various histologies in an effort to grow tumor-infiltrating lymphocytes (TIL) using flasks and subsequent expansion in semipermeable bags. The five most prevalent tumor types from which TIL were successfully initiated were melanoma (25 successful initiates in 34 tumor samples, 74% success rate), colorectal cancer (12 of 18, 67%), renal cell carcinoma (9 of 12, 75%), breast (4 of 5, 80%), and sarcoma (5 of 7, 71%). The overall success rate for all tumors was 67 of 94 (71%). There were no instances of contamination from the time of culture initiation through harvesting of the final cell product for clinical use. The mean number of days to reach successful initiation (> 5 x 10(8) cells) was 35 +/- 24 days (mean +/- SD). TIL were then expanded from these successful initiates for either a repeated low-dose therapy (TIL reinfusion numbers of 5 x 10(8)-5 x 10(9) or for a repeated high-dose therapy (> 5 x 10(9)-5 x 10(10). The mean number of days to expand a TIL culture from the time of initiation to treatment for a first low-dose TIL was 59 days (range, 27-94 days) compared with 80 days (range, 33-209 days) for high-dose TIL. For patients who received a second or third high-dose TIL treatment, the average number of days needed to expand TIL was 39 days (n = 10) if there was no intervening cryopreservation of TIL, compared with 49 days (n = 10) if the culture had to be reestablished from cryopreserved TIL. For patients who received a second or third low-dose TIL, the mean number of days needed to expand TIL was 23 days (n = 3) if there was no intervening cryopreservation compared with 42 days (n = 17) if cultures had to be reestablished after cryopreservation of TIL. Low-dose TIL displayed predominantly CD4+ phenotype in 76% of 42 cultures, whereas high-dose TIL displayed predominantly CD8+ phenotype in 84% of 44 cultures. Cells bearing the natural killer (NK) phenotype (CD3-, CD56+) and the lymphokine activated killer (LAK) phenotype (CD3+, CD56+) were present in both low- and high-dose TIL cultures, but these phenotypes were never predominant. Cytotoxicity testing consistently demonstrated the persistence of NK and LAK activity in addition to the killing of allogeneic and autologous melanoma tumor targets. This work confirms that TIL cultures from most tumor types can be successfully established and expanded for therapeutic use, and repeated expansion from continuous TIL culture or cryopreserved TIL for repeated treatments is feasible. Such cultures are predominantly T lymphocytes that are phenotypically heterogeneous, and these phenotypes do not remain constant during prolonged time in culture.

Tumor localization by tumor infiltrating lymphocytes labeled with indium-111 in patients with metastatic renal cell carcinoma, melanoma, and colorectal cancer.

BACKGROUND: Adoptive immunotherapy with autologous tumor infiltrating lymphocytes (TIL) is a promising approach for cancer bio-therapy. One issue, however, is whether such cells actually migrate to sites of tumor after intravenous infusion. There have been several reports of tumor uptake of radiolabeled TIL in patients with metastatic melanoma, but efforts to visualize tumor with radiolabeled TIL in other tumor types reportedly have been unsuccessful. METHODS: Eight patients with metastatic cancer (5 renal, 2 melanoma, 1 colon) received an intravenous infusion of 2 to 100 billion autologous TIL, including 50 million TIL which had been conjugated to 500 microCi Indium-111, co-administered with interleukin-2 (IL-2). One patient received 1 gm/m2 of cyclophosphamide one day prior to TIL; seven patients received interferon alpha 2b for 4 days prior to receiving TIL. Total body gamma camera imaging, including single photon emission computerized tomography (SPECT), was performed at 24 and 48 hours. RESULTS: All eight patients had demonstrable uptake of 111-Indium-labeled TIL into one or more known sites of tumor. There were no known sites of tumor which were not imaged. Metastatic sites imaged included bone, brain, mediastinal and perihilar lymph nodes, lung and liver parenchyma, abdominal periaortic nodes, and a pelvic mass. One patient served as a negative control in that the TIL scan was negative at a time when she had no evident disease, but a few weeks later had a positive TIL scan which lead to a diagnosis of axillary recurrence. CONCLUSION: Uptake of radiolabeled TIL, whether CD8+ or CD4+, by metastatic renal cell carcinoma and other carcinomas was similar to that previously reported in melanoma. Pretreatment with cyclophosphamide was not a prerequisite for imaging, and TIL uptake did not predict tumor response.