Search

To the Editor: Since our group first reported on the potential clinical applications of the human tumor stem-cell, or clonogenic, assay in the Journal,1 we think some comment from us is warranted regarding the recent article by Selby et al.2 and the editorial by Von Hoff.3 We clearly agree that it is important to plate good single-cell suspensions, that drug-sensitivity criteria are still in the developmental stages,4 that some drugs may require the evaluation of various exposure times, that inappropriate drug concentrations may be misleading, that drugs requiring bioactivation present special problems, that radiation-survival curves with plateaus at. © 1983, Massachusetts Medical Society. All rights reserved.

Differential Staining Cytotoxicity (DiSC) assay is the prototype for a closely related family of assays based on the concept of total cell kill, or, in other words, cell death occurring in the entire population of tumor cells. It is probably the most versatile of the cell-death end points, in that it (1) can be applied to both solid and hematologic neoplasms, (2) can be applied to specimens in which it is not possible to obtain a pure population of highly enriched tumor cells, and (3) can be applied to a wide variety of drugs, ranging from traditional cytotoxic agents to biological response modifiers with activity mediated through tumor-infiltrating effector cells, to "targeted" kinase inhibitors, and to antivascular agents, such as bevacizumab and pazopanib. The basic principles of the assay are to culture three-dimensional fresh tumor cell clusters in anchorage-independent conditions. At the conclusion of the culture period, Fast Green dye is added to the microwells, the contents of which are then sedimented onto permanent Cytospin centrifuge slides and then counterstained with hematoxylin-eosin or Wright-Giemsa. "Living" cells stain with the cytologic stain in question and can be identified as either normal or neoplastic, based on standard morphologic criteria. "Dead" cells stain blue-green. Nonviable endothelial cells appear as strikingly hyperchromatic, blue-black, and often refractile objects, which may be readily distinguished from other types of dead cells. This assay has been biologically and clinically validated in a number of ways, as described in this chapter.

Background: Angiogenesis studies are limited by the clinical relevance of laboratory model systems. We developed a new method for measuring dead microvascular (MV) cells in clinical tissue, fluid and blood specimens, and applied this system to make several potentially novel observations relating to cancer pharmacology., Methods: Dead MV cells tend to have a hyperchromatic, refractile quality, further enhanced during the process of staining with Fast Green and counterstaining with either haematoxylin-eosin or Wright-Giemsa. We used this system to quantify the relative degree of direct antitumour versus anti-MV effects of cisplatin, erlotinib, imatinib, sorafenib, sunitinib, gefitinib and bevacizumab., Results: Bevacizumab had striking anti-MV effects and minimal antitumour effects; cisplatin had striking antitumour effects and minimal anti-MV effects. The ;nib' drugs had mixed antitumour and anti-MV effects. Anti-MV effects of erlotinib and gefitinib were equal to those of sunitinib and sorafenib. There was no detectable VEGF in culture medium without cells; tumour cells secreted copious VEGF, reduced to undetectable levels by bevacizumab, greatly reduced by cytotoxic levels of cisplatin + anguidine, and variably reduced by DMSO and/or ethanol. We observed anti-MV additivity between bevacizumab and other drugs on an individual patient basis. Peripheral blood specimens had numerous MV cells which were strikingly visualized for quantification with public domain image analysis software using bevacizumab essentially as an imaging reagent., Conclusions: This system could be adapted for simple, inexpensive and sensitive/specific detection of tissue and circulating MV cells in a variety of neoplastic and non-neoplastic conditions, and for drug development and individualized cancer treatment.

The generation of pluripotent stem cells from an individual patient would enable the large-scale production of the cell types affected by that patient's disease. These cells could in turn be used for disease modeling, drug discovery, and eventually autologous cell replacement therapies. Although recent studies have demonstrated the reprogramming of human fibroblasts to a pluripotent state, it remains unclear whether these induced pluripotent stem (iPS) cells can be produced directly from elderly patients with chronic disease. We have generated iPS cells from an 82-year-old woman diagnosed with a familial form of amyotrophic lateral sclerosis (ALS). These patient-specific iPS cells possess properties of embryonic stem cells and were successfully directed to differentiate into motor neurons, the cell type destroyed in ALS.

In the opinion of the authors, good medical practice demands the avoidance of ineffective therapies and good cancer chemotherapy generally demands the avoidance of inactive drugs. The authors present their viewpoint that randomized trials are not required to prove that physicians should not administer inactive drugs and that existing cell culture assays are acceptably accurate in identifying inactive drugs. They describe and advocate the use of two such tests to aid in choosing between different forms of therapy that would otherwise be equally acceptable in the absence of test information. Strengths, weaknesses, and applications of the assays are discussed.

Tumor-specific cytotoxicity was measured in fresh human biopsy specimens by a modification of the differential staining cytotoxicity assay. ImuVert, a cytokine inducer derived from Serratia marcescens, which produces broad-spectrum activation of both macrophages and lymphocytes, was dramatically more effective when it was tested in tumors obtained from patients with previously treated, chemotherapy-responsive adenocarcinomas (breast and ovary) than when it was tested in tumors obtained from either previously untreated patients or previously treated patients with chemotherapy-refractory adenocarcinomas (colon, lung, pancreas, stomach, kidney, gallbladder, uterus, and prostate). Similar findings, relating to prior chemotherapy treatment status, were obtained for tumor necrosis factor and interferon gamma, but not for interleukin-2 or interferon alpha. On the basis of these findings and on other evidence in the literature, we speculate that response to chemotherapy produces massive release and processing of tumor antigens. We further speculate that this response leads to a state in which the human immune system is primed (via in situ vaccination) to respond to exogenous macrophage-activation signals with potent, specific antitumor effects.

Bayes' theorem has been used to describe the relationship between the accuracy of a predictive test (posttest probability) and the overall incidence of what is being tested (pretest probability). Bayes' theorem indicates that laboratory assays will be accurate in the prediction of clinical drug resistance in tumors with high overall response rates (e.g., previously untreated breast cancer) only when the assays are extremely (greater than 98%) specific for drug resistance. We developed a highly specific drug-resistance assay in which human tumor colonies were cultured in soft agar and drugs were tested at high concentrations for long exposure times. Coefficients for concentration x time exceeded those reported in contemporaneous studies by about 100-fold. We reviewed 450 correlations between assay results and clinical response over an 8-year period. Results were analyzed by subsets, including different tumor histologies, single agents, and drug combinations. Extreme drug resistance (an assay result greater than or equal to SD below the median) was identified with greater than 99% specificity. Only one of 127 patients with tumors showing extreme drug resistance responded to chemotherapy. This negligible posttest probability of response was independent of pretest (expected) probability of response. Once this population of patients with tumors showing extreme drug resistance had been identified, posttest response probabilities for the remaining cohorts of patients varied according to both assay results and pretest response probabilities, precisely according to predictions based on Bayes' theorem. This finding allowed the construction of a nomogram for determining assay-predicted probability of response.

Ethacrynic acid [2,3-dichloro-4-(2-methylene-1-oxobutyl)phenoxyl] acetic acid, is a water-soluble diuretic agent that has been shown to potentiate the in vitro cytotoxicity of chemotherapeutic agents in established cell lines. We used the differential staining cytotoxicity (DiSC) assay to determine whether ethacrynic acid at 1 and 3.3 microM would potentiate the cytotoxicity of nitrogen mustard and/or doxorubicin in primary cultures of hematologic neoplasms from heavily pretreated patients and in normal peripheral blood lymphocytes. At 3.3 microM, ethacrynic acid was toxic to 8 of 24 (33%) tumor specimens studied. In subsequent studies, ethacrynic acid at 1 microM was toxic to only 2 of 54 (4%) tumor specimens. Significant enhancement for doxorubicin or nitrogen mustard was confined to lymphatic malignancies and to normal peripheral blood lymphocytes. Interspecimen variability was observed, with no enhancement in most individual specimens, 2-fold enhancement in some specimens, and 4-fold enhancement in occasional specimens. Clinical trials will be required to determine whether the observed in vitro activity for ethacrynic acid is associated with clinical benefit in unselected or assay-selected patients.

Clinically significant hemorrhage immediately after percutaneous liver biopsy is an uncommon, although well-recognized risk of the procedure. However, hemorrhage from the biopsy site delayed a day or more after the procedure is a seldom-appreciated potential complication. Delayed hemorrhage from the liver biopsy site may go unsuspected; in fact, the patient may no longer be under direct medical surveillance. We describe two cases of such delayed hemorrhage which we attribute to intrahepatic hematoma at the site of biopsy. Neither patient had a prebiopsy contraindication to the procedure. One patient suffered a fatal intraperitoneal hemorrhage 15 days after liver biopsy, the other a massive, but nonfatal intrahepatic hemorrhage 41 hours after the procedure. In the first patient, a hemorrhagic diathesis developed in the immediate period after biopsy; no predisposition to hemorrhage could be identified in the latter patient, even retrospectively. The admittedly rare possibility of delayed hemorrhage should be considered whenever a liver biopsy is performed.

We tested the ability of the differential staining cytotoxicity (DiSC) assay to discriminate between sensitive and resistant cell populations in human lymphatic neoplasms. First, the in vitro activity spectra of the most important drugs paralleled the known clinical activity spectra of the same agents. Second, there were highly significant correlations between in vitro chemosensitivity and the results of clinical chemotherapy. Third, specimens from previously untreated patients were significantly more sensitive to the most important drugs than were specimens from patients who had previously received chemotherapy. Finally, metachronous assays performed on specimens from the same patients showed little change in chemosensitivity if there had been no intervening chemotherapy between the times that the first and second assays were performed. However, if the patients had received intervening chemotherapy between the times of the first and second assays, the specimens in the second assays tended to be significantly more resistant than were the specimens in the first assays. These data indicate that the DiSC assay may be of value in the design of strategies to circumvent drug resistance in human lymphatic neoplasms.

One hundred nineteen patients were entered onto a randomized trial of the role of intravenous hyperalimentation (IVH) in patients with small-cell lung cancer. IVH was given during the first 30 days of induction chemotherapy to 54 patients. IVH did not effect any improvement in response or survival from therapy. In view of the lack of benefits from IVH, an analysis was made of the toxicities suffered by the 54 patients receiving IVH as well as any effects IVH might have made on chemotherapy-induced toxicity. Toxicities observed included mechanical difficulties with the catheter leading to temporary or permanent discontinuation of the IVH (11 patients), subclavian vein thrombosis (one patient), sepsis in nine patients v none of the 62 control patients, fluid overload (27 patients), hyponatremia (25 patients), and hyperglycemia requiring insulin (13 patients). Patients receiving IVH had higher granulocyte counts on days 14 and 21 of the first cycle of chemotherapy. Analysis shows that this difference is likely caused by fever and infection associated with IVH rather than any nutritional effect on granulopoiesis. In this population of patients, IVH had significant complications but did not ameliorate chemotherapy-induced toxicity and it did not effect any clinical benefit. Future studies of adjunctive nutritional therapy must consider the significant risk in this older population and must limit IVH volume or exclude patients with even mild compromise in cardiovascular functions. Further, any new trial must have a significant rationale for adjunctive use to justify the potential risks.

We need practical laboratory methods for predicting the chemosensitivity of human neoplasms. Over the past 20 years, numerous investigators have implied or stated with increasing certitude that clonogenic assays are the most valid (or only valid) approach to predictive chemosensitivity testing. We feel that this point of view may have insufficient scientific validity and may be harmful to progress in this area. In addition to well-known technical problems, there are serious theoretical problems with clonogenic assays. These include the disruption of normal cell-to-cell interactions, the possibility that true tumor stem cells may be largely nondividing (G0) cells, while cells forming colonies are exclusively dividing cells, the possibility that clonogenic cells may largely represent cells which are not true stem cells, and the fact that clonogenic assays have the ability to measure cell kill over a narrow log range, while meaningful clinical responses require a multiple-log cell kill. This latter fact mandates the use of unrealistically low drug concentrations to avoid excessive false-positive results. Neither theoretical concepts, direct experimental data, nor clinical correlations support the alleged superiority of clonogenic assays. Clonogenic assays may not be advantageous compared to other more practical methods of estimating the general chemosensitivity of proliferating cells. In contrast, there is a growing body of literature which indicates that early evidence of cell damage in the entire tumor cell population may accurately predict for a multiple-log stem cell kill and meaningful clinical response. Future studies should continue to develop and test assays based on alternative methods for detecting cell kill in the proliferating and total tumor cell populations.

Two hemagglutinating components purified from the lima bean and composed of identical 62,000-molecular-weight subunits have mitogenic activity in normal and leukemic human lymphocyte cultures. Component II (247,000 molecular weight), a tetramer with four saccharide binding sites, is severalfold more active than component III (124,000 molecular weight), a dimer with two binding sites per molecule, as a mitogen for normal lymphocytes. Component II also stimulates mitogenesis in populations of cultured leukemic lymphocytes and appears to have greater activity than phytohemagglutinin-P in leukemic lymphocyte cultures obtained from some patients. Lima bean component III has minimal mitogenic activity toward leukemic lymphocytes. Components II and III appear to compete for the same cell-surface binding sites, but component II has a much greater capacity than component III to trigger DNA synthesis at low concentrations (2.5-25 mug/ml) in normal lymphocyte cultures. These results suggest that the mitogenic activity of lima bean lectin components is directly related to their valence.

The chemosensitivity of fresh specimens of human transitional cell carcinoma was tested using a novel dye exclusion assay based upon Fast Green staining, cytocentrifugation, and H & E counterstaining. Assays were successful (three or more drugs tested) in 22/24 specimens obtained by cytoscopy and 2/2 specimens obtained by open biopsy. Three of 17 specimens had more than 70% cell kill after a 1-hour exposure to thiotepa (500 micrograms/ml), compared with 10/18 after exposure to mitomycin C (250 micrograms/ml), and 19/19 after exposure to doxorubicin (100 micrograms/ml). There was a trend for specimens with high histologic grades or low control viabilities in short-term culture to be more sensitive to mitomycin C than specimens with low histologic grades or high control viabilities. Comparison of these data with published clinical data suggests that intrinsic chemosensitivity may be the most important determinate of clinical response to intravesical chemotherapy with thiotepa, while drug penetration to the tumor cells may primarily determine clinical response to intravesical doxorubicin. Thiotepa may be most valuable in treating large lesions which are intrinsically sensitive, while doxorubicin may be valuable in nearly all small lesions. Improved drug distribution to tumor sites could improve intravesical therapy with doxorubicin.

Fresh specimens of human lymphatic neoplasms were tested with the differential staining cytotoxicity assay. Cells from relapsed patients with acute lymphoblastic leukemia (ALL) were significantly more resistant to vincristine, dexamethasone, and doxorubicin in the assay than were cells from previously untreated patients. The putative C kinase inhibitors verapamil (V), imipramine (I), lidocaine (L), tamoxifen (T), chlorpromazine (C), and haloperidol (H) were then tested singly, in combination with each other (VILTCH, ITCH, and VL), and in combination with vincristine. At concentrations judged to be clinically achievable, VILTCH itself was occasionally toxic to ALL and chronic lymphocytic leukemia. The VILTCH combination clearly potentiated the cytotoxic activity of vincristine in five of eight ALL specimens from relapsed patients and potentiated vincristine in 18 of 30 chronic lymphocytic leukemia specimens. It also potentiated vincristine in two of six specimens of multiple myeloma and five of six specimens of non-Hodgkin's lymphoma. The VILTCH combination had no significant effects in fresh cultures of normal human lymphocytes. The most active drugs in the VILTCH combination appeared to be verapamil and lidocaine. We conclude that the differential staining cytotoxicity assay is a useful tool to study the circumvention of clinically acquired drug resistance. While the mechanism of the observed enhancement of the cytotoxic effects of vincristine is not known, it is possible that combinations of putative C kinase inhibitors may reduce drug resistance in human lymphatic neoplasms.

The effect of menadiol (vitamin K3) on fresh specimens of human lymphatic neoplasms (HLN) was tested by means of the differential staining cytotoxicity assay. Menadiol was tested alone and in combination with standard antineoplastic agents. Drug effects were then compared with the effects of the same drugs in normal human lymphocytes and in fresh specimens of human non-small cell lung cancer. By itself, menadiol was moderately toxic to HLN, but not to normal lymphocytes or non-small cell lung cancer. Menadiol, menadione, and two structurally related congeners were equitoxic to HLN cells, but sodium metabisulfite (present in menadiol solutions as a preservative) was nontoxic. Menadiol increased the cytotoxic effects of a number of standard agents in HLN but not in normal lymphocytes. Cell survival times with mechlorethamine, vincristine, and dexamethasone were converted from a range characteristic of drug resistance (ie, range observed in relapsed patients) to a range characteristic of drug sensitivity (ie, range observed in untreated patients) in the presence of menadiol. These effects occurred at a concentration (2.0 micrograms/ml; 4.7 microM) of menadiol which is probably clinically achievable and which did not deplete intracellular glutathione. Menadiol should receive clinical testing as a chemosensitizing agent in HLN.

Cells from bronchoscopy washings from 30 patients with lung cancer were plated in tissue culture using a human tumor stem cell assay technique. These culture results were subsequently compared to routine cytology readings carried out on the same specimens. The stem cell culture showed colony growth from 9 of 11 cytologically positive bronchoscopy washing specimens. More colonies grew from cytologically positive specimens (average of six colonies per plate) than from cytologically negative specimens (average of 0.8 colonies per plate). With additional refinements, the stem cell culture system could be used to improve the diagnostic yield of bronchoscopy washings. However, at present, the low number of tumor colonies grown from the washings precludes performing chemosensitivity studies on tumor cells from the bronchoscopy washing specimens.

Dissociated cancer cells are exposed to antineoplastic drugs (5 X 10(4) viable cells/drug for 1 hr or continuously) and cultured for 4 to 6 days in liquid medium. Cells are then stained with Fast green dye, sedimented onto slides with a Cytospin centrifuge, and counterstained with a modified hematoxylin and eosin technique. Dead cells stain with Fast green, and living cells stain with hematoxylin and eosin. Cell kill is calculated as percentage of control based on the relative numbers of living tumor cells, living non-tumor cells, and dead cells. Drug sensitivity could be assayed in 125 of 162 specimens of human neoplasms obtained from malignant effusions (16 of 18), excisional biopsies (31 of 44), needle biopsies (34 of 47), endoscopic biopsies (18 of 23), peripheral blood samples (19 of 20), and bone marrow aspirates (five of seven). The assays were successful (median of ten drugs tested) in: 46 of 64 adenocarcinomas; four of 11 squamous cell carcinomas; five of seven lymphomas; six of seven melanomas; two of four sarcomas; 18 of 20 transitional-cell carcinomas; 14 of 15 small-cell carcinomas; seven of eight myelomas; 12 of 12 chronic lymphocytic leukemias; seven of nine acute leukemias, and four of five "undifferentiated" carcinomas. The assay results demonstrated a strong correlation between the in vitro chemosensitivity of different types of tumors and the known clinical response patterns of these tumors. This assay can be used to determine which specific cells are killed in a heterogeneous cell population. Further work is needed to determine if this assay may be useful in blind screening trials for antineoplastic agents or if it may be of clinical use in predicting response to agents which are not cycle specific.

The following factors must be considered when dye exclusion assays are interpreted. (a) It may require several days for lethally damaged cells to lose their membrane integrity following a cytotoxic insult. (b) During this time, the "surviving" cells may continue to proliferate. (c) Also during this time, some lethally damaged cells may undergo an early disintegration, so that they are not present to be stained with dye at the end of the culture period. Factors b and c may cause an underestimate of cell kill when the results of the assay are based upon the traditional "percent viability" expression. In order to overcome these problems, an internal standard was developed and tested. This was based upon the addition of a constant number of permanently fixed duck erythrocytes to the cultures of cells from two different established tumor cell lines. Results were based upon comparisons of the ratios of "viable" tumor cells to duck erythrocytes on permanent cytocentrifuge slides prepared from the cultures. This novel "ratio" method was found to be a more sensitive index of drug-induced cell kill than the traditional percent viability method. A standard agar cloning assay gave somewhat higher estimates of cell kill than the ratio method, although both assays were in qualitative agreement for the drugs tested. All three assays demonstrated a clear dose-effect relationship for most of the drugs tested. Dye exclusion assays may have a useful role in chemosensitivity testing in vitro.