Ariel C. Hollinshead

Separation of Skin Reactive Intestinal Cancer Antigen from the Carcinoembryonic Antigen of Gold

Soluble fractions of human intestinal cancer and fetal intestinal cell membranes produced delayed hypersensitivity reactions in patients with intestinal cancer. These soluble fractions and perchloric acid extracts of intestinal cancer cells were fractionated by polacrylamide-gel electrophoresis. The Gold carcinoembryonic antigen was found in a region of the gels different from that of the skin reactive antigen.

Soluble Membrane Antigens of Lip and Cervical Carcinomas: Reactivity with Antibody for Herpesvirus Nonvirion Antigens

With the use of antibody for herpesvirus nonvirion antigens (not structural components of the virus) complement fixing reactivity has been shown for soluble membrane antigens separated from lip and cervical carcinomas but not for similar extracts from normal vaginal tissue or intestinal carcinoma. Neither the serum obtained from the guinea pig before hyperimmunization with the herpesvirus nonvirion antigen nor the antiserum of guinea pigs immunized with comparable uninfected cell extracts reacted with these tumor soluble membrane antigens. Since the above soluble membrane antigens could be specific markers for the presence of virus genome within the tumor cells, the findings could support an etiological role of herpesvirus in selected human malignancies.

SKIN-REACTIVE SOLUBLE ANTIGEN FROM INTESTINAL CANCER-CELL-MEMBRANES AND RELATIONSHIP TO CARCINOEMBRYONIC ANTIGENS

Abstract Delayed hypersensitivity reactions in seventeen of nineteen patients with carcinoma of the colon and rectum were elicited by soluble fraction obtained from the membranes of autologous tumour cells. Negative reactions were obtained with comparable fractions obtained from normal cells. Skin-reactive antigen was also detected in the digestive-tract cells of first-trimester and second-trimester fetuses. Carcinoembryonic antigen (C.E.A.) was detected in many of the fractions producing positive skin reactions. The skin-reactive tumour antigen appeared to be closely related, and possibly identical, to the C.E.A. but, although found in comparable fractions, it may well be distinct from the C.E.A.

Soluble membrane antigens of human malignant melanoma cells.

Delayed hypersensitivity reactions to soluble components of the cell membranes of autologous and allogeneic tumors were elicited in a series of patients with malignant melanoma. Two skin reactive antigens were prepared using stepwise low frequency sonication of cell membranes, chromatography of membrane sonicates on Sephadex G‐200, and separation by polyacrylamide gel electro‐phoresis (PAGE). One group of antigens, in Sephadex fraction II, PAGE region a, appears to be melanoma‐associated antigen. Sephadex fraction II, and further separated Sephadex fraction II PAGE region a, produced no reactions in 21 of 22 tests in patients with cancers other than melanoma; positive reactions were seen in 17 of 22 patients with early stage melanoma and in 7 of 19 patients with late stage melanoma. The other antigen, from Sephadex fraction III, PAGE region b, was also reactive in 5 of 6 early stage breast cancer patients. Comparable separated proteins from Sephadex fraction III, and Sephadex fraction III PAGE region b, of normal black skin cell membrane sonicates were reactive in 4 of 9 early stage melanoma patients but were negative in 7 late stage melanoma patients. Sephadex fraction III and Sephadex fraction III PAGE region b produced no reactions in patients with renal, cervical, colonic, and head and neck cancer; 9 of 21 patients with early stage melanomas were skin test positive; positive reactions were seen in 13 of 18 patients with late stage melanoma.

Adjuvant specific active lung cancer immunotherapy trials. Tumor-associated antigens.

The 10‐year cumulative experiences of five year survivals of patients entered into a successful phase II specific active tumor‐associated antigen (TAA) immunotherapy trial, a successful phase III specific active immunotherapy trial A and of patients from centers with acceptable protocol violation levels of an unsuccessful specific active immunotherapy trial B are evaluated. Here the authors report the efficacy of specific active TAA immunotherapy when the protocol is adhered to strictly, where the induction of cell‐mediated immunity to TAA indicated a successful adherence to the protocol rather than the strategic result when centers from the third trial with major violations are included. The authors repeat here a summary of each of the three separate trials, each of the three trials having been reported elsewhere in their entirety, so that these total results may be compared to the present analysis. The survival experiences of a total of 234 lung cancer Stage I and Stage II patients, including all violations, from centers in northern New York, northern New Jersey, western Pennsylvania and eastern Canada show a statistically valid (P = 0.0002) 5‐year survival difference between the control groups (receiving adjuvant alone or no treatment) at 49% survival and the specific active immunotherapy groups at 69% survival. The best promise of specific active immunotherapy alone in an adjunctive treatment setting is with early stage lung cancer. In addition to tests which monitor the effect of TAA immunotherapy induction of long‐lasting cell‐mediated immunity, tests (monoclonal antibody‐derived epitope enzyme immunoassays) were developed to monitor specific, early antibody rises in the bloodstream (circulating humoral immunity).

Specific active immunotherapy in patients with adenocarcinoma of the colon utilizing tumor‐associated antigens (TAA). A phase I clinical trial

Twenty‐two patients received specific active immunotherapy (TAA vaccine once per month for 3 months), with the duration of follow‐up, as of July 1984, ranging from 3 months to 36 months (median, 21 months). Of these, seven had Dukes B2, seven had Dukes C, and eight had Dukes D lesions. All received surgical resection, and those with Dukes D disease underwent resection of all metastases where possible, with six clinically disease‐free at the time of initiation of therapy. The age range of the 22 patients was 40 to 73 years (median, 60 years); sex distribution was 12 males and 10 females. All patients were monitored by physical examination and by laboratory parameters including complete blood count, liver and renal function tests, blood chemistries, urinalysis, chest x‐ray, carcinoembryonic antigen levels, migration inhibition assays, complete immune complexes, serum chemistries, helper and suppressor and total T‐cell and B‐cell assays, and TAA antibody levels. As measured by delayed cutaneous hypersensitivity skin test and by migration inhibition assays (MIA), a strong postimmunization response is developed approximately 5 months after vaccination is completed. There were no clinical or biochemical manifestations of any type of systemic toxicity including hepatic, renal, gastrointestinal, respiratory, or neurologic during the period of follow‐up. All patients developed skin ulcers at the vaccination and required 4 to 5 months to heal. With this small number of patients in a Phase I trial, survival is indicative of the safety of the vaccine only: 82% of the patients are alive (mean survival, 21 months) thus far, and 59% of the patients are without evidence of disease (NED) (mean NED, 22 months). These studies, therefore, justify a Phase II‐III trial in a larger number of patients and have provided selection of appropriate monitoring tests for the larger trial.

Soluble membrane antigens of brain tumors. I. Controlled testing for cell-mediated immune responses in a long surviving glioblastoma multiforme patient

A patient with glioblastoma multiforme survived 18 years after diagnosis and underwent 20 operations for extracranial metastasis. An immunologic survey of the patient was made over a 1‐year‐period using in vitro tests of lymphocyte responsiveness and skin tests with control and tumor antigens isolated from autologous and allogenic brain cell membranes. Two tissue‐associated soluble cell membrane antigens also present in normal white matter, and two tumorassociated antigens (TAA) produce cell‐mediated immune responses in patients with brain tumors. One of these tumor‐associated antigens predominates in meningioma cells. In addition some low molecular weight components appeared, which seemed to be unique for the glioblastoma cells from the longsurviving patient. Cancer 40:660–666, 1977.

Pilot studies using melanoma tumor-associated antigens (TAA) in specific-active immunochemotherapy of malignant melanoma

Highly purified melanoma TAA which induce melanoma‐related cellmediated immune responses have been further characterized using hyperimmune TAA antisera after affinity chromatography for double immunodiffusion‐immunoelectrophoresis and indirect immunofluorescence studies. An additional study of antigenic modulation was performed in 23 nonanergic and seven anergic melanoma patients, tested simultaneously with melanoma TAA prepared from primary and metastatic tumors, which had been obtained from one patient at different time periods. The results of pilot clinical trials are reported, including toxicity, timing and dosage studies in 20 patients and subsequent studies of patients with metastatic melanoma treated at three separate centers, using a single lot of purified, allogeneic melanoma TAA. The results of these latter studies in 51 patients with Stage III (distantly metastatic) melanoma and in five patients with earlier stages of disease indicate that: (1) when the interval from primary therapy to recurrence is greater than one year and when liver, bone and brain are not involved, partial or total clinical regression may be noted in up to 25% of patients with metastatic disease receiving immunochemotherapy; (2) when total regression does occur, the effect usually lasts from one to three years; (3) cytoreductive (debulking) surgery, when possible, in cutaneous, nodal retroperitoneal, and visceral regions may enhance the response to specific active immunochemotherapy, although some debulked patients had less tumor burden and this factor alone may lead to an improved prognosis in patients undergoing any subsequent treatment; (4) when circulating inhibitory factors are modified through preimmunization chemotherapy, an enhanced host response may be seen; and (5) Cancer Serum Indices (CSI) may be useful in predicting recurrence and in following tumor load and response to therapy. Information obtained from these studies suggests the need for further trials to determine the effect of immunization on patients with earlier stages of disease where recurrence rates remain high, and to evaluate the mechanisms of tumor rejection or tumor progression in the face of immune stimulation.

Specific active lung cancer immunotherapy. Immune correlates of clinical responses and an update of immunotherapy trials evaluations.

The mechanisms of action of the specific active immunotherapy of solid tumors have not been defined. In an attempt to characterize some of these mechanisms, we report controlled studies of humoral immune responses and cell‐mediated immune (CMI) responses in lung cancer patients with Stage I and Stage II adenocarcinoma and squamous cell cancer receiving pure tumor‐associated antigen (TAA) specific active immunotherapy or combination immunochemotherapy. At 5 to 6 months postimmunotherapy, the humoral immune response measurements are predictive of response to therapy/survival in early lung cancer patients, permitting decisions as to whether to continue therapy. Patients with adenocarcinoma respond to combination chemoimmunotherapy by showing stronger or earlier responses to tests of immunity. Cell‐mediated immunity to TAA at 17 to 24 months was far greater in patients receiving immunotherapy or immunochemotherapy compared with control patients, and also correlated with early humoral immune response and with 5‐year survival. Here we report a further subset analysis of Stage I and Stage II lung cancer patients in a successful Phase III US specific active immunotherapy trial as substantiating the experience with Stage I patients in a successful Phase II Canadian trial. We analyze failures and suggest additional therapies, especially a chemoimmunotherapy trial indicated by our analyses of humorocellular immune variables reported here.

Adjuvant specific immunotherapy of resectable squamous cell lung carcinoma. Analysis at the eighth year.

SummaryFrom June 1976 to June 1981, 86 patients with resectable (Stage I and II) squamous cell lung carcinoma were entered into a randomized controlled study with three arms:I.Control Group — no treatment postoperatively.II.Specific Immunotherapy Group — three monthly doses of 500 μg of tumor associated antigen (TAA) emulsified with complete Freunds adjuvant (CFA).III.Nonspecific Immunotherapy Group — three monthly doses of CFA emulsified in saline. All the patients in the study received skin tests with PPD (5TU) and 100 μg of the same TAA used for the immunotherapy at 1, 4, 6, 9, and 12 months postoperatively.Patients in both immunotherapy groups showed a tendency for a better disease-free interval and overall survival compared to those of the control, but these interval and beneficial therapeutic effects were statistically significant only in the Group III patients who had no hilar lymph node metastasis (T1N0 and T2N0).Although Group III was originally designated as a nonspecific immunotherapy group, retrospectively, it should be called a lowdose specific immunotherapy group because these patients actually received a total of 500 μg of TAA (as skin tests) and three doses of CFA at separate sites.