mechanism of interferon action in hairy cell leukemia: a model … · hairy cell leukemia is a...

(CANCER RESEARCH 52, 1056-1066, March 1, 1992)

Review

Mechanism of Interferon Action in Hairy Cell Leukemia: A Model of EffectiveCancer Biotherapy1

Suresh Vedantham,2 Haim Gamliel, and Harvey M. Golomb

Section of Hematology/Oncology, Department of Medicine, University of Chicago Medical Center, Pritzker School of Medicine, Chicago, Illinois 60637

Abstract

Hairy cell leukemia (HCL) is one of the few malignancies for whicha-interferon (IFNa) is considered effective first-line therapy. However,

the mechanisms of action of this agent in vivo have been the subject ofmuch debate; in particular, the issue of whether clinical improvement inIFNa-treated HCL patients is dependent upon enhancement of hostdefenses or upon direct actions of IFNa upon the hairy cell remainsunresolved. In this review, we examine the evidence supporting both linesof argument and synthesize this information within the framework ofclinical studies of IFNa in HCL, the purpose being to determine whichproposed mechanisms of IFNa action are indeed effective in vivo. Fromour analysis, it appears that the beneficial effects of IFNa upon immunefunction are important in decreasing the frequency of infectious complications of HCL but that these effects are probably not responsible forhairy cell elimination and cannot therefore account for major responsesto IFNa therapy. We conclude that the primary mechanism of action ofIFNa in HCL involves the induction of hairy cell differentiation towardsa stage less responsive to growth factor stimulation, the primary consequence being proliferativi- inhibition. These effects may mimic events

that occur during normal lymphocyte development, suggesting that thebenefits of biotherapeutic agents might best be harnessed via studies ofthe effects of multiple and sequential biological response modifiers uponthe growth and differentiation patterns of normal and malignant cells.Hairy cell leukemia could thus serve as an excellent model in which toinvestigate combined cancer biotherapy; the implications of our presentunderstanding of IFNa in HCL to the biotherapy of cancer are discussed.

Introduction

The general approach to the study of cancer treatment haschanged significantly in recent years. Whereas in the pastemphasis was placed upon searching for better cytotoxic chem-

otherapeutic drugs, we now appreciate the fundamental vulnerabilities of such agents and have refocused much of our attention to more natural modalities. Research into the diverseeffects of cytokines has been particularly fruitful, and we nowpossess greatly improved insight into the events involved in theregulation of immune response. The discovery that these samemolecules may have direct effects upon neoplastic cells causedmuch excitement and gave rise to the concept of cancer biotherapy. Advances in recombinant DNA technology and thesubsequent mass production of such agents rendered this ideapractically feasible, and the 1980s marked the emergence ofsuch factors in clinical trials, as lone agents, and in chemother-

apeutic protocols. Agents thought to be most promising are

Received 10/31/91 ; accepted 12/17/91.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' This work was supported in part by the HÃ©matologieResearch Foundation

and the Hairy Cell Leukemia Research Foundation.2To whom requests for reprints should be addressed, at 729 East Third St.,

Hinsdale, IL 60521.

IFNa,3 IL-2, and TNF. Although all three are presently in

volved in clinical trials, IFNa is the only cytokine to be approved by the Food and Drug Administration as first-linetherapy for cancer (1).

a-Interferon has been tested in vitro and in vivo against awide variety of malignancies, including common solid tumors,melanoma, sarcomas, lymphomas, and leukemias. In vitro studies and early trials were extremely encouraging, and IFNa wastouted as a potential cure for many cancers. IFNa has shownsome effectiveness in patients with metastatic melanoma (2),bladder carcinoma (3), renal cell carcinoma (4), multiple myeloma (5), some forms of non-Hodgkin's lymphoma (6), and

CML (7, 8). However, clinical trials over the last 5 years havenot yielded the spectacular results that were earlier anticipated,and the long-term prognosis for most IFNa-treated cancerpatients remains essentially unchanged. As a result, one nowperceives within the medical community a leveling in the enthusiasm for IFNa as an antineoplastic agent.

We feel that this is premature and that further studies arerequired to justify such pessimism. One critical problem is thatalthough much is known about the biological effects and biochemical reactions associated with IFNa in vitro, little is understood about which of these events are the clinically significantmechanisms of action. Until this vital information is acquired,we will be unable to assess the full potential of interferon incancer; it might be that adjunctive therapy could be designed toaugment the specific actions of IFNa in certain cancers andthereby maximize its effectiveness.

HCL is one of the few malignancies for which IFNa hasassumed a prominent therapeutic role (9). However, muchcontroversy still exists concerning the mechanism of interferonaction in HCL. Because the treatment of HCL with IFNacurrently represents the best example of successful cancer biotherapy, increasing attention has been paid in recent literatureto answering this question. The major goals of this review are:(a) to describe the effects of IFNa upon immune function inHCL patients; (b) to delineate the specific actions of IFNaupon the individual hairy cell and to determine whether agoverning principle such as cellular differentiation can be invoked to explain them; and (c) to correlate this analysis with invivo studies of IFNa action and to thus deduce which proposedmechanisms of action are actually involved in the clinical response to IFNa. Such knowledge could be utilized to providefurther direction in IFNa research and to design methods ofoptimizing the effects of IFNa in HCL treatment. These principles could in turn be applied to the challenge of improvingcytokine therapy in other malignancies.

3The abbreviations used are: IFNa, a-interferon; IL-2, interleukin 2; TNF,

tumor necrosis factor; CML, chronic myelogenous leukemia; HCL, hairy cellleukemia; CLL, chronic lymphocytic leukemia; sIL-2R, soluble interleukin 2receptor; IL-2R, interleukin 2 receptor, NK, natural killer, FACS-fluorescence-activated cell sorting; I.Ak, lymphokine-activated killer; BCGF, B cell growthfactor.

1056

Research. on December 12, 2020. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

MECHANISM OF INTERFERON IN HAIRY CELL LEUKEMIA

Hairy Cell Leukemia and Its Treatment with Interferon

Hairy cell leukemia is a chronic lymphoproliferative disorderthat is characterized clinically by splenomegaly, pancytopenia,and the presence of a malignant clone of tartrate-resistant acidphosphatase-positive cells (10-12) displaying unique morphological features in the peripheral blood and bone marrow. Theorigin of the hairy cell has been the subject of much debate.Although evidence for monocytic origin has been presented (13,14) and T-cell variants described (15), the bulk of the evidencefavors a B-cell origin for these neoplastic cells. Hairy cellsexpress surface immunoglobulin (16) and other B-cell-associ-ated surface antigens (17), and molecular biological studieshave revealed the presence of immunoglobulin gene rearrangements in these cells (18). Studies with monoclonal antibodies and differentiation inducers indicate that hairy cellsprobably stem from preplasmatic B-cells, and HCL is thereforeclassified as a B-cell malignancy slightly more differentiatedthan CLL (19, 20). HCL, like CLL, runs a chronic course, withmedian survival being over 5 years from the time of diagnosisprior to the introduction of systemic therapy (21).

Although treatment options for HCL have included chemotherapy, radiation, leukapheresis, androgens, and bone marrowtransplantation (22), at present the primary recommended therapies are splenectomy, IFNa, and pentostatin (in clinical trials).Splenectomy results in the alleviation of symptomatic splenomegaly, and approximately 50% of patients undergoing splenectomy achieve normalization of peripheral blood counts (22).However, splenectomy has no effect upon bone marrow disease,and most patients eventually require additional therapy; he-matological parameters and the degree of bone marrow cellu-larity are now utilized to identify subsets of patients who mightderive long-term benefit from the procedure (23). The varioussystemic treatment modalities used thus far have had seriouslimitations and have generally failed to produce durable responses. Clinical trials involvingpentostatin have demonstratedoverall response rates of greater than 95%, of which 75% havebeen complete responses (24-26). However, the long-term tox-icities of this agent have not yet been determined, and inparticular, reports of delayed immunosuppression in HCL patients receiving pentostatin raise concerns that must be resolvedbefore this drug can be used as first-line therapy in HCL patients(27). Recently, durable complete responses with minimal tox-icity were reported to occur in 11 of 12 HCL patients treatedwith 2-chlorodeoxyadenosine (28). These results appear to bequite promising, although the safety and efficacy of this drugin HCL must be further evaluated in a larger series of patients.

At present, IFNa represents the primary systemic therapyfor patients with HCL. Most HCL patients treated with IFNashow a favorable clinical response to the drug, overall responserates being approximately 80-90%, with a 5% complete response rate (21). It has now been established that the clinicalresponse to IFNa differs from that generally seen with chemo-therapeutic agents. A prompt response to standard chemotherapy is usually seen if treatment has been effective. In contrast,the response to IFNa may develop slowly, and even partialresponses may be quite durable with continuing treatment.Recovery of platelet count can occur within 2 weeks, andreduction of splenic and peripheral blood tumor burden generally occurs within 2 months. Neutropenia resolves in 2-3months, and erythrocyte and neutrophil counts may continueto increase for up to 12 months (21,22). Bone marrow improvement generally occurs over a period of 2-6 months, although

eradication of the disease does not seem to occur, and relapseoccurs in most patients after discontinuation of IFNa. Currentrecommendations are for 12 months of therapy (21); the mediantime to relapse after discontinuation of IFNa is approximately2 years (29). IFNa is successful in reinducing remission in mostof these patients. These results suggest that despite its lowcomplete response rate, IFNa therapy may lead to lastingclinical remissions more consistently than have previous formsof treatment. In addition, IFNa, a biological agent, exhibits aside effect profile which compares very favorably with those ofstandard chemotherapeutic agents, including pentostatin. Unlike pentostatin, IFNa can be utilized safely in patients withactive infections. Consequently, IFNa is presently the treatmentof choice for most HCL patients in need of systemic therapy(21).

Two general mechanisms have been proposed to explain theability of IFNa to induce remissions in HCL patients: enhancement of host immune responses; and direct antiproliferativeeffects upon the hairy cell.

Effects of Interferon upon Immune Function

a-Interferon has been demonstrated to exert effects at multiple levels in the generation of immune response: it influencesthe production and differentiation of hematopoietic elements;enhances their recognition of foreign antigens; and augmentsthe ability of effector cells to destroy their selected targets. Therelevance of such actions to the biotherapy of HCL has beenthe subject of intense debate. In particular, the issue of whetherII No-enhanced immune function mediates hairy cell elimination is central to our understanding of the actions of IFNa inHCL.

Effects upon the Hematological Profile. One of the primaryclinical manifestations of HCL is the marked pancytopenia thatoften afflicts its victims, predisposing them to symptomaticanemia, bleeding complications, and recurrent infections. Inparticular, the profound degree of monocytopenia found inHCL patients has long been recognized as a prominent featureof the disease (30). Severe impairments in natural killer cellactivity have been noted (31), and defects in T-lymphocyteimmunity have also been reported, rendering HCL patientsmore susceptible to infections with atypical organisms as wellas common bacterial pathogens (32). Infectious complicationsrepresent the leading cause of death in patients with HCL (21).

It has long been believedthat the cytopenias observed in HCLare the result of splenic sequestration of hematopoietic elementsand bone marrow infiltration by hairy cells. Indeed, splenectomy has been shown to result in improvement in at least onehematological parameter in 90% of cases (21). However, thealleviation of pancytopenia is often incomplete, and even patients responding well to splenectomy eventually experiencerecurrence of cytopenias. It has therefore been suggested thatthe pancytopenia in HCL is due primarily to impaired production of hematopoietic elements rather than to hypersplenism(12). This may in part be due to space-occupying bone marrowinfiltration by leukemic cells; however, recent evidence favors arole for hairy cell-derived substances that inhibit the growthand differentiation of specific hematopoietic cell types.

Most investigations of such phenomena have focused uponthe presence of sIL-2R in the sera of patients with HCL. It haslong been known that the malignant cells in most HCL casesexpress the CD25 (Tac) antigen (18), which is thought torepresent a determinant on the interleukin 2 receptor. This

1057




molecule is normally expressed on activated T-lymphocytes andon a subset of activated B-lymphocytes (33); in these cell types,the IL-2R mediates responsiveness to IL-2 and thus plays animportant role in cellular proliferation.

In 1985, the release of a soluble form of the IL-2R fromactivated human lymphocytes in vitro was first reported (34).Soon afterwards, the sera of eight untreated HCL patients weretested, and it was discovered that all eight samples containedhigh levels of the sIL-2R (35). Other investigators soon confirmed these findings in larger series of patients (36, 37). It wastherefore suggested that the sIL-2R might play a role in theimpairment of T-lymphocyte and natural killer immunity observed in HCL patients by competetively inhibiting the bindingof IL-2 to its receptors on T- and NK cells, depriving them ofan important activation/proliferation stimulus. In addition totheir role in preventing infections, T-lymphocytes and NK cellsproduce a multitude of cytokine growth factors that serve toregulate bone marrow stem cell function. Hence, a paucity ofthese elements might disrupt critical balances in the normalhematopoietic environment and further contribute topancytopenia.

Various leukemia-associated bone marrow inhibitors havebeen characterized over the last 15 years (38, 39), but untilrecently, attempts to isolate hairy cell-derived factors that might

mediate hematopoietic stem cell suppression had not met withsuccess (40). During the last few years, several groups haveisolated activities from the sera of HCL patients that inhibit invitro colony formation by bone marrow stem cells. Two groupsreported on factors the inhibitory activity of which appeared tobe primarily directed against myeloid and erythroid progenitorcells (41, 42). One was shown to be a protein of M, 5000-6000the other is presently awaiting molecular characterization. Inaddition, it was recently reported that bone marrow sera fromHCL patients contain a tumor necrosis factor that causessignificant progenitor cell suppression in colony-forming assays(43). Although further characterization of these factors is necessary, their presence might explain the defect in monocyticdifferentiation seen in HCL patients. Adverse consequencesassociated with monocytopenia may include reductions in phag-ocytic activity and antigen-presenting ability, along with decreased synthesis of important monokines such as interleukin1 and IFNa. Deficiencies in these compounds might causefurther suppression of multiple cell lineages.

The beneficial effects of low-dose IFNa upon the hematolog-ical profile of patients with HCL have been firmly established(22). Patients responding to IFNa with significantly diminishedhairy cell burden often begin to experience alleviation of cyto-penias before hairy cell percentages in the bone marrow startto decrease (44). Furthermore, it has been demonstrated thateven patients not achieving complete or partial bone marrowresponses display significant improvement in bone marrowfunction, as is evidenced by decreased infection rates and bloodproduct transfusion requirements (45). It is thus clear that theeffects of IFNa upon peripheral counts occur partly via pathways that do not depend upon reduction of tumor load andfibrosis.

a-Interferon is known to exert effects upon multiple cell typesinvolved in normal immune response. In particular, it has beenshown to promote cellular activation and differentiation in cellsof several hematopoietic lineages, including monocytes (46).Because of the profound monocytopenia seen in HCL, it hasbeen postulated that a state of endogenous IFNa deficiencyexists in these patients. Indeed, IFNa biotherapy in HCL has

been likened to insulin "replacement" therapy in diabetes mel-

litus (47), this theory holding that exogenous IFNa may initiallycompensate for the lack of endogenous IFNa, resulting inresumption of normal IFNa-dependent hematological processes, such as monocytic differentiation. These effects mightoccur quite early in the course of therapy and may in partexplain why HCL patients experience clinical improvementprior to reduction in the hairy cell index.

Decreased helper/suppressor T-cell ratios have been noted inHCL patients (48), consistent with observed deficiencies in cell-mediated immunity. Elevated levels of soluble CDS have beendiscovered in the sera of these patients (49); this finding isthought in part to signify increased suppressor cell activity.IFNa has been shown to drastically reduce soluble CDS levelsas well as the number of CD8+ T-cells in the peripheral blood.It has been shown that this reduction, with the resultant restoration of normal helper/suppressor ratios (50), correlates wellwith the alleviation of bone marrow suppression seen in IFNa-treated HCL patients (49).

A third route by which IFNa improves peripheral bloodparameters may involve interruption of humoral pathways ofhairy cell-mediated bone marrow suppression. One group reported that the sera of patients responding well to IFNa contained significantly less stem cell-inhibitory activity than those

of untreated patients (42). These reductions correlated wellwith decreases in peripheral blood hairy cell counts and maytherefore have reflected changes in tumor load and factor-secreting capacity. The issue of whether IFNa causes changesin the production of this inhibitor before reductions in hairycell index are seen is at present unclear. Studies concerning thefunction of other hairy cell-derived factors during IFNa therapyhave not yet emerged.

A fourth major mechanism by which IFNa may improveimmune function in HCL patients revolves around its ability todecrease serum levels of soluble interleukin 2 receptor. In theoriginal study on eight HCL patients with high sIL-2R levels,it was reported that IFNa treatment resulted in steadily decreasing serum sIL-2R levels throughout the 12-month course oftherapy in all patients tested. A strong correlation between sIL-2R levels and clinical status was seen, and a possible role forthis parameter in monitoring response to IFNa therapy wassuggested (35). Other groups have since confirmed and extended these observations (36). In studies at our institution(37), a statistically significant correlation between sIL-2R levelsand hairy cell index was found, demonstrating that much of thereduction in sIL-2R level may be due to the decrease in tumorload seen with IFNa treatment. Diminished levels of sIL-2Rmay result in exposure of sensitized cells to increased levels ofavailable IL-2 and may thus enhance their ability to clonallyproliferate in response to antigenic challenge. Further studiesare required to determine if levels of sIL-2R can be correlatedwith susceptibility of HCL patients to typical and atypicalinfections.

At present, the mechanisms underlying the hematologicalimprovement seen in IFNa-treated HCL patients have not yetbeen fully elucidated. However, it is clear that these actionsencompass alterations in the production of a wide range of celltypes, including monocytes, neutrophils, erythrocytes, platelets,and various lymphocyte subsets. Such changes may result inrestoration of essential humoral and cellular balances, affordingthe body a more appropriate hematopoietic milieu in which tocontinue synthesis of important immune elements. IFNa mayalso improve immune response by increasing the availability of

1058




cytokines essential for antigen-dependent cellular proliferation.Significant reductions in infection frequency attest to the clinical significance of these effects.

Effects upon Antigen Recognition. Recognition of foreignantigens by the human immune system is mediated by productsof the major histocompatibility complex (HLA gene locus).Class I antigens, expressed in cells from most human tissues,are thought to target malignant and virus-infected cells fordestruction by CD8+ cytotoxic T-lymphocytes. Class II antigens, limited mainly to B-lymphocytes, monocytes, and macrophages, are involved in the presentation of foreign or autoan-tigens to CD4+ helper T-cells, and the subsequent developmentof antigen-specific clones of T-lymphocytes and antibody-producing B-lymphocytes (51). HLA antigens are expressed uponmalignant cells in a wide variety of human and animal neoplasms; it is hypothesized that they may play an important rolein the recognition of tumor cells by the immune system (52,53). Malignant cells from most HCL patients express HLA-ABC (Class I) and HLA-DR (Class II) antigens, consistent withthe B-cell origin of these cells (17).

a-Interferon has been demonstrated to modulate expression

of HLA determinants on many normal and malignant cell types.It has been shown to enhance Class I determinant expressionon B-lymphocytes (54), on monocytes, and on tumor cells fromvarious malignancies, including Burkitt's lymphoma and CML

(55). Class II determinant expression on B-lymphocytes andmonocytes is generally not affected by IFNa (54); however,IFNa has been shown to induce and modulate Class II antigenexpression on neoplastic cells from CML (55) and metastaticmelanoma (56), cancers in which IFNa has demonstrated sometherapeutic value.

In 1986, FACS analysis was used to show that hairy cellscultured with IFNa demonstrate significantly increased reactivity with the OKDR, Leu-10, and OK la monoclonal antibodies(57). Further studies utilizing cytofluorography and immunoe-lectron microscopy confirmed these findings and reported thatthe observed enhancements in HLA Class II expression occurin HCL patients responding to IFNa therapy but not in non-

responders or in control cells from normal individuals andpatients with other B-type leukemias (58-60). However, it hasnot been shown that this is a primary cause of favorable responses to IFNa therapy; the finding of increased HLA expression may merely indicate that IFNa exerts direct effects uponthe hairy cells of responders and that increased HLA expressionis a marker of hairy cell sensitivity to antiproliferative effectsof IFNa. Further studies are required to clarify the issue ofwhether changes in Class II HLA antigen expression renderhairy cells more susceptible to recognition by immune elements.

Another question which remains to be answered involves theexpression of Class I HLA antigens upon normal and malignantcells in HCL patients. IFNa has been reported to cause increased expression of Class I determinants on hairy cells (60).However, this finding may be of little functional significance,inasmuch as it has been repeatedly demonstrated that hairycells are poor targets for major histocompatibility complex-restricted cytotoxic T-cell lysis and that this mechanism isprobably not an important feature of the immune defenseagainst HCL (61, 62). Nevertheless, the ability of IFNa toinduce Class I HLA antigens on normal B-lymphocytes andmonocytes in HCL patients may result in improvements inthe antigen-presenting ability of monocytes or in increased sensitivity of B-cells to T-cell-derived signals. This may serve asanother mechanism by which IFNa decreases the frequency of

infectious complications in HCL patients.Effects upon Effector Cell Cytotoxicity. NK cells have long

been considered likely mediators of antitumor activity (63).They are thought to function as important elements in thenonspecific defense system, accomplishing non-major histocompatibility complex-dependent lysis of abnormal or neoplastic cells. Severely deficient NK cell function has been widelyreported in HCL patients (64, 65); it was thought that thisimpairment might contribute to disease progression and anincreased predisposition to infections. Mechanisms underlyingthis deficit may include decreased production of NK cell precursors due to HCL-induced bone marrow suppression, deficient activation of such cells due to soluble IL-2R-mediatedreductions in available 11,2. target binding defects (65), andimpairment of cytolytic mechanisms.

a-Interferon has been shown to augment NK cell cytotoxicityagainst a variety of neoplastic cells, including those from severalhematological malignancies (66, 67). IFNa exerts beneficialeffects upon the NK system of HCL patients; whether theseactions are essential to the elimination of hairy cells has beenan issue of controversy. IFNa has been reported to increase thenumber of large granular lymphocytes, precursors of NK cells,in the peripheral blood of HCL patients (64), although conflicting reports exist concerning changes in peripheral blood NKcells (49, 68). Studies have clearly demonstrated significantincreases in the cytolytic capabilities of NK cells against K562tumor cell targets after in vivo IFNa treatment of HCL (64,69). One group recently reported improvement in NK-K562cell binding, increased killing ability of bound targets, andincreased release of natural killer cytotoxic factor by NK cellspreincubated with IFNa (70). Several studies have found correlations between improvement of NK function, hematologicalimprovement, and decreases in sIL-2R levels and hairy cellindex in IFNa-treated patients (60, 69, 71, 72). However,improvements in NK function tend to occur late in the courseof IFNa therapy and have not been shown to precede hematological recovery and tumor load reduction (69, 72).

The most compelling evidence speaking against the role ofNK cells in hairy cell killing comes from studies which haveclearly demonstrated that hairy cells are not susceptible to lysisby NK cells despite induction with IFNa (64, 73, 74). Lack ofhairy cell recognition and binding by NK cells has been implicated as a cause for this resistance (61, 65). Hairy cells are alsoresistant to lysis by IFNo-preincubated LAK cells (75). It istherefore highly unlikely that NK/LAK-mediated cytolysisplays an important role in the elimination of malignant cells inHCL. However, IFNa-boosted enhancement of NK/LAK cy

totoxicity may improve the deficient immune function observedin HCL patients and may thus contribute to the reduced frequency of opportunistic infections that is seen after IFNatherapy.

Summary: Interferon and Immune Function in Hairy CellLeukemia. The effects of IFNa upon the immune system inpatients with HCL are quite complex and involve actions atmultiple levels of immune function: production of hemato-poietic elements; recognition of foreign antigens; and killing ofselected targets. One issue critical to our understanding of themechanisms of action of IFNa in HCL concerns the extent towhich changes in hairy cell index can be explained by IFNa-induced augmentation of host immune function. Our review ofthe literature leads us to conclude that the existing evidencedoes not support a major role for IFNa-mediated enhancementof immune function in the elimination of hairy cells and that

1059




these actions cannot account for the dramatic responses toIFNa therapy seen in HCL patients. Because IFNa does notcompletely eradicate hairy cells from the bone marrow, themajor goal of therapy is not necessarily to cure the disease butrather to prolong quality survival, in part by reducing thefrequency and severity of HCL-associated complications. Improved host defenses in IFNa-treated HCL patients do appearto prevent opportunistic infections, the leading mortality-associated complication of HCL; these effects may be critical duringthe early stages of therapy, in which the full effects of IFNaupon the hairy cell index have not yet occurred. Hence, IFNo-induced improvements in immune function play an importantrole in the treatment of HCL.

Direct Effects of Interferon upon the Hairy Cell

HCL patients responding well to IFNa therapy demonstratedramatic reductions in hairy cell burden in both the peripheralblood and bone marrow (21). Because existing evidence doesnot support the possibility of significant hairy cell eliminationby IFNa-enhanced host defenses, an explanation for the observed improvement may be rooted in direct effects of IFNaupon the hairy cell. IFNs have been shown to inhibit theproliferation of various normal and malignant cell types (76,77) and have even been associated with phenotypic reversion oftransformed cells (78, 79). The delayed timing of IFNa-inducedcytoreduction in HCL, the paucity of complete responses toIFNa therapy, and the apparent failure of IFNa to completelyeradicate hairy cells from the bone marrow serve to dismiss theidea of IFNa as a cytotoxic agent in HCL. Rather, it is thoughtthat IFNa functions as a cytostatic agent in HCL, inhibitinghairy cell proliferation and thereby accomplishing decreases intumor burden.

Although experiments in recent years have enabled us todescribe in some detail the diverse effects of IFNa upon thehairy cell, many questions arise regarding the relevance of theseindependent actions to hairy cell function and to the clinicalimprovement seen in HCL patients. Can these changes beunderstood in the context of normal B-lymphocyte development? Why is IFNa so effective in HCL but not in CLL andother related lymphoproliferative disorders? We support theview that IFNa-induced hairy cell differentiation accounts formost of the observed changes in hairy cells exposed to IFNaand for the clinical efficacy of this agent in HCL. The characteristics of IFNa-treated hairy cells include a more maturephenotype and an impaired response to growth factors. Themajor consequence is a reduction in proliferative capacity.

Effects upon Response to Growth Factors

Leukemic cells are believed to exhibit phenotypes whichcorrespond to discrete stages in hematopoietic differentiation.It has therefore been postulated that these cells may respond tothe same cytokine growth factors as their nonmalignant counterparts. Experiments utilizing differentiation inducers in conjunction with monoclonal antibodies have demonstrated thathairy cells probably represent malignant partially activated B-lymphocytes, slightly more mature than the neoplastic "resting"

B-cells of chronic lymphocytic leukemia (19, 20). Normal B-cells at this level of maturity are beginning the process of clonalproliferation and eventual differentiation into antibody-secreting plasma cells; hence, they are quite responsive to a varietyof biological response modifiers. It has therefore been proposed

that these same molecules might control the proliferation ofhairy cells in vivo.

B Cell Growth Factor. Hairy cells in vitro and in vivo demonstrate a low proliferative index in comparison to other malignant cells, even when stimulated with standard B-cell mitogens,and have thus been difficult to maintain in long-term culture.However, the addition of BCGF to cultured hairy cells has beenshown to stimulate hairy cell growth (80), as illustrated byincreased thymidine incorporation and by the maintenance ofcontinuous hairy cell lines. As a result, it was suggested thatBCGF might contribute to hairy cell proliferation in vivo. In1987, cytosolic extracts from hairy cells of untreated patientswere found to contain a BCGF-like activity when tested onhairy cells as well as on normal BCGF-dependent human B-cell lines. Chromatographie analysis indicated that this activitywas similar to the activity of normal T-cell-derived cytosolicBCGF. This study suggested that hairy cells contain, and insome cases secrete, an autostimulatory growth factor, implicating autocrine growth mechanisms in the pathophysiology ofHCL (81).

In 1986, it was first demonstrated that IFNa inhibits BCGF-induced hairy cell growth using six cultured cell lines (82). Thisfinding was quickly confirmed by other investigators, who observed similar phenomena in hairy cells taken from IFNa-treated patients (83). These studies were extremely important,because they represented the first proofs of the direct antipro-liferative actions of IFNa on hairy cells. Taken in conjunctionwith the report of HCL-derived BCGF, they suggested that themechanism of IFNa action in HCL may involve inhibition ofautocrine loops that serve to promote hairy cell proliferation.

Interleukin 2. The expression of the IL-2 receptor on subsetsof activated B-cells was a source of confusion until it was shownthat IL-2 may play a direct role in B-cell proliferation and

differentiation (84,85). The presence of the CD25 (Tac) antigenon hairy cells supports the concept that hairy cells are neoplasticcounterparts of CD25+ partially activated B-cells. IL-2 alonehas been shown to exert only minor proliferative effects uponhairy cells and does not appear to possess the ability to maintainhairy cells in culture (80). However, it has been demonstratedthat treatment of activated B-lymphocytes with BCGF greatlyenhances the surface expression of IL-2R (86); immunoelectronmicroscopic studies in our laboratory have extended these observations to include hairy cells, revealing significant enhancement of CD25 expression in BCGF-treated hairy cells (87).These findings are compatible with the notion that BCGFrecruits normal and malignant B-cells to become responsive toIL-2 by increasing their surface expression of the IL-2 receptor.

As discussed earlier, the sera of patients with HCL have beenfound to contain elevated levels of soluble IL-2R, and IFNahas been shown to greatly reduce these levels (35). Studies havesince demonstrated strong correlations between IFNa-inducedsIL-2R reduction and clinical response to therapy (36). Acorrelation between serum sIL-2R level and hairy cell index hasbeen noted (37); however, it is not clear to what extent reductions in sIL-2R simply reflect decreases in tumor burden, ratherthan decreased expression of IL-2R on hairy cells. In 1989,

FACS analysis was used to show decreased CD25 expressionon hairy cells cultured with IFNa (62). Our immunoultrastruc-tural studies have confirmed these observations via examinationof individual hairy cells exhibiting persistently diminishedexpression of CD25 after 3 days or more of culture with IFNa(87). In addition, IFNa consistently abrogated BCGF-inducedincreases in CD25 expression on hairy cells in these experi-

1060




ments. It is therefore likely that part of the action of IFNÂ«inHCL involves decreasing surface IL-2R expression, renderinghairy cells immune to the proliferative effects of IL-2.

The strong correlations found between decreases in sIL-2R

level and clinical response to IFNÂ« therapy reflect severalongoing processes, one of which is probably tumor load reduction. However, it is now quite clear that IFNÂ« also causesdecreases in IL-2R expression on individual hairy cells. On theone hand, this reduction may simply reflect alterations in hairycell function due to antiproliferative effects of IFNa, unrelatedto IL-2. Alternatively, decreases in responsiveness to IL-2 mayin part mediate the antiproliferative effects of IFNÂ«on hairycells. We believe that BCGF-treated hairy cells represent aunique IL-2-sensitive stage of B-cell development, that promotion of further differentiation by IFNa results in diminishedsensitivity to BCGF and IL-2, and that these changes are in

large part responsible for the clinical response to IFNa.Tumor Necrosis Factor. TNF is a monocyte-derived cytokine

that mediates diverse effects as a modulator of nonspecificimmune response (88). Although clinical investigations of TNFhave focused upon its role as a cytotoxic agent against varioushuman cancers (89), it has also been demonstrated to stimulatethe proliferation of several normal and malignant cell types (90,91). Hairy cells have been reported to express surface receptorsfor TNF (92), and treatment of activated hairy cells with TNFhas been shown to elicit a proliferative response (93). In addition, hairy cells from some patients are thought to secrete TNF;evidence for this is derived from analysis of HCL culturesupernatants and from the discovery of TNF in the bone marrow sera of untreated HCL patients (94). Treatment of hairycells with TNF results in greatly enhanced cytosolic interleukin1, interleukin 6, and TNF mRNA and protein levels (95).Hence, the TNF system clearly fulfills the criteria for an auto-

crine loop.a-Interferon has been shown to inhibit TNF-dependent hairy

cell growth in experiments by several investigators (93, 95).This effect appears to be independent of changes in surfaceTNF receptors. Following IFNÂ« administration to culturedhairy cells, increased levels of cytosolic 2-5 A synthetase mRNAare noted; soon afterward, decreases in mRNA levels for theabove cytokines are seen (95). It has therefore been suggestedthat IFNÂ«might alter hairy cell proliferation via degradationof cytokine mRNA. In view of the demonstrated TNF autocrineloop inhibition by IFNa in vitro, it is quite possible that diminution of TNF production by IFNÂ«plays a role in mediatingthe antiproliferative effects of IFNa in vivo.

The above discussion illustrates the importance of B-cellgrowth-promoting agents in the pathophysiology of HCL.

Other factors, including interleukin 4, interleukin 5, and interleukin 6, have been reported to alter hairy cell growth capabilities (93, 96), but these interactions have not yet been wellcharacterized. Further studies are required to determine theextent to which hairy cell growth is dependent upon each of theabove factors and to fully assess the consequences of inhibitionby IFNa. Such experiments will need to control for the possiblepresence of multiple cytokines in the supernatants of hairy cellcultures. The discovery that IFNÂ«may operate via the inhibition of autocrine and paracrine growth factor loops is indeedfascinating and suggests that further exploration into growthfactor-stimulated proliferation of other tumor cell types may be

necessary for cancer biotherapy to succeed.

Effects upon Oncogene Expression

Aberrant protooncogene expression is thought to contributeto the expression of transformed phenotypes in animal andhuman malignancies. Normal and malignant hematopoieticcells have often been used in investigations into these phenomena; as a result, much is known about oncogenic function inleukemic cells. In particular, the expression of the c-myc and c-

fos genes has been intensively studied in the neoplastic cells oflymphoproliferative disorders. As a result, much has beenlearned about the nuclear regulation of lymphocyte proliferation, activation, and differentiation.

Cell cycle-dependent transient expression of the c-myc andc-fos genes has been shown to occur during the activation of

human lymphocytes (97, 98). These genes encode nuclear proteins that are thought to be necessary for commitment to DNAreplication and cell division in response to growth factors. Ithas been suggested that these proteins may activate cells toprogress from a quiescent phase of the cell cycle (G0) into d(97) and thus accelerate their transition into S phase. Expression of either gene throughout the cell cycle might result in aloss of proliferative control due to an inability of the cells toremain in Go.

It has been demonstrated that levels of c-myc mRNA fall

dramatically following the induction of differentiation of various cell types (99, 100). In 1984, it was found that treatmentof various neoplastic lymphoid cells with type I interferonsresulted in G0/Gi arrest in 10 of 11 cases in which proliferationwas inhibited (101). In 1986, endogenous IFN/3-induced c-myc

suppression was shown to be a normal event during the differentiation of hematopoietic cells, and this reduction was seen tocorrelate with cellular arrest at the G0/G, transition into thecell cycle (102). Selective reductions in c-myc mRNA in IFN/3-treated Daudi-Burkitt's lymphoma cells were reported, and it

was suggested that this might play a role in the inhibition ofproliferation (103). These observations were soon confirmedand extended to include IFNa (104). Other studies have sinceshown that IFNÂ« induces terminal differentiation of leukemiccells in part by reducing the half-life of c-myc mRNA (105,

106). Hence, IFNÂ« is thought to play a major role in modulatingprotooncogene expression in normal and malignant lymphoidcells, the consequences of which include regulation of cellularproliferation and differentiation.

In 1986, the expression of the c-fos and c-myc genes in cells

taken from HCL patients before and during IFNa therapy wasexamined (107). c-fos mRNA was detected in the hairy cells

from nine of nine untreated patients, and IFNÂ« was shown tomodulate c-fos expression in both responding and nonrespond-ing patients. However, c-myc transcripts were not detected intypical HCL cells but were present in both variant (nonrespond-ing) cases tested. In these cases, IFNa failed to modulate c-myc

expression even after 6 weeks of administration. The lack ofexpression of c-myc in typical HCL cases is consistent with the

low proliferative index of hairy cells. The fact that IFNa successfully modulated c-fos production in these cases may indicate

that IFNÂ« does affect hairy cell function at the level of oncogeneexpression. It may be that this regulation is sufficient to inhibitproliferation in typical HCL cases but that c-myc expression invariant cases is capable of overriding the c-fos system, resulting

in continued proliferation and a clinically resistant phenotype.The authors of this paper suggested that the failure to modulatec-myc expression might indicate the limits of low-dose IFNa

therapy for HCL. In resistant patients, the failure of IFNa to

1061




reduce c-myc levels might prevent the G0/Gi arrest and subsequent differentiation seen in other IFNa-treated cell types.Further studies are needed to delineate the cellular effectsassociated with c-fos expression in hairy cells and the relationship between the modulation of this gene by IFNo- and growthfactor-dependent control of the cell cycle.

Effects upon Hairy Cell Differentiation

The impressive responses of HCL patients to low doses ofan apparently cytostatic agent compel one to consider mechanisms of action that mimic the physiological effects of IFNaupon normal cellular function. It has been proposed that theprimary effect of IFNa in HCL is to induce hairy cells todifferentiate into mature cells less responsive to proliferativesignals. The rationale behind this theory is derived from thedemonstration of the ability of IFNa to induce differentiationin a variety of normal and malignant cells, including leukemiccells of both lymphoid and myeloid origin (108-110). Evidencesupporting the role of IFNa as a differentiation agent in HCLcomes from numerous immunological and ultrastructural studies which clearly demonstrate phenotypic alterations in hairycells treated with IFNa.

Most investigations of hairy cell differentiation have utilizedmonoclonal antibodies in conjunction with standard inumino-

fluorescence methods or FACS to analyze surface expressionof B-lymphocyte antigens. Early studies using these techniquesled many researchers to conclude that IFNa-induced alterationsin hairy cells are not accompanied by phenotypic maturation.However, in most of these experiments, only percentages ofpositive cells were examined; recent FACS data using modalfluorescence in combination with percentage positivity havereported different results, illustrating that percentage positivityand measurements of fluorescence per cell are not necessarilyconcordant. Methodologies such as modal FACS and Â¡mmunoscanning electron microscopy are thus ideally suited to suchinvestigations and may provide more accurate information regarding antigenic regulation at the single cell level.

A second problem lay in the interpretation of the results ofearly immunological studies. Because of the designation ofHCL as a B-type lymphoproliferative disorder, assessments ofhairy cell differentiation were often based upon measurementsof surface and cytoplasmic immunoglobulin and plasma cellmarkers. However, it has since been shown that although hairycells are mature B-cells that often express the plasma cellmarker PCA-1(111), their patterns of differentiation differ significantly from those of normal B-lymphocytes. Treatment ofhairy cells with phorbol esters to induce differentiation in vitrohas been demonstrated to endow them with macrophage-likequalities; they adhere to plastic dishes and acquire ultrastructural properties similar to those of macrophages (112-115).The explanation for this strange transformation is at presentunknown; it may be that the underlying pathophysiology ofHCL involves an irreversible departure from the classical B celldifferentiation program or that hairy cells require specific maturation factors in order to display plasmacytic features (116).Whether such factors capable of restoring typical B cell differentiation tendencies to hairy cells indeed exist is presentlyunknown. In addition, it has not been shown that terminal hairycell differentiation is required for growth suppression to occur;hence, the inability of IFNÂ«to cause terminal differentiationas measured by changes in immunoglobulin production andplasma cell antigen expression does not exclude partial hairy

cell differentiation as a cause of proliferative inhibition. LikeB-cells, maturing hairy cells lose positivity for many B-cell andhairy cell surface antigens as they progress. As a result, recentstudies utilizing immunological methods to examine hairy celldevelopment have been extremely valuable in characterizingIFNa as a differentiation stimulus for hairy cells.

Class II HLA antigens are expressed differentially during B-cell development and are thus regarded as differentiation markers in normal and leukemic B-lymphocytes (117). In particular,expression of these antigens and the CD25 (Tac) antigen hasbeen associated with the activation of normal B-lymphocytes(33). The ability of phorbol esters to induce Tac expression onB-type CLL cells (118), to significantly enhance the expressionof Class II HLA on CLL cells (119), and to provide them withthe classical surface morphology of hairy cells (118) supportsthe reliability of these surface markers as indicators of malignant B-cell activation and differentiation.

Experiments in various laboratories have clearly shown thatIFNa enhances the expression of Class II HLA antigens onhairy cells. In 1986, increased HLA-DR expression was reported to occur after incubation of samples from three HCLpatients with IFNa in vitro (57). These findings were confirmedby others using various methods to compare antigenic expression in HCL patients who responded favorably to IFNa withpatients refractory to therapy. In one study, leukemic cellsexhibited a definite increase in HLA density and a slight decrease in transferrin receptors in responding patients but not inrefractory patients. The authors suggested that the enhancedHLA expression indicated progression towards differentiationand that the decrease in transferrin receptors was correlatedwith growth reduction (58). In 1988, IFNa-induced increasesin HLA-DR expression on hairy cells were shown to be relatedto favorable clinical course (60). More recently, changes inhairy cell surface immunoglobulin type without production ofcytoplasmic immunoglobulin or increased expression of plasmacell markers were demonstrated to occur following IFNa therapy (116). It was suggested that these changes were consistentwith the induction of partial differentiation but that furthercofactors might be required to completely overcome the maturation arrest seen in the malignant hairy cells.

In 1989, a quantitative FACS study was performed uponsplenic and peripheral blood cells taken from six HCL patients.A large panel of monoclonal antibodies was used, and bothpercentage positivity and modal fluorescence were utilized toanalyze phenotypic alterations in hairy cells exposed in vitro toIFNa. The results agreed with those of previous investigators,demonstrating significantly enhanced HLA antigen expressionon IFNa-treated hairy cells. Enhanced CD22 expression alongwith decreased staining for CD25, FMC7, surface immunoglobulin light chains, CD 19, CD9, and HC2 were also found.It was proposed that the enhancement of HLA Class II andCD22 expression and the reductions observed in expression ofthe other antigens might be explained by activation/partialdifferentiation of the hairy cells, because these antigens tend tobe lost during the later stages of B-cell differentiation. Treatment of hairy cells with phorbol esters resulted in similarantigenic changes, supporting this interpretation of the data.However, the IFNa-treated hairy cells did not become completely negative for CD 19, CD22, and HLA Class II; this wastaken to indicate a failure of the hairy cells to complete differentiation (62).

Our own studies involving the immunoscanning electronmicroscopy analysis of antigenic and ultrastructural changes on

1062




IFNa-exposed hairy cells have allowed us to confirm that the Concluding Remarksconclusions reached by previous investigators are indeed validat the level of the single hairy cell. We observed significantlyincreased HLA-DR expression and decreased CD25 expressionon hairy cells cultured for 3 days with IFNa and further noticeda significant decrease in CD lie expression after 5 days ofculture with IFNa (87). In addition, we described specificultrastructural alterations in hairy cells treated with IFNa. Upto one-third of the cells displayed significant membrane changesincluding the appearance of a bud-like formation at one poleand a "bubbling" membrane covering the cell body and/or the

surface ruffles. Redistribution of antigens upon the cell surfacewas also noted. These morphological effects were observed onlyupon cells displaying increases in Class II HLA expression anddecreases in CD25 expression after 3 days of culture with IFNa.The HCL patients from whom these cells were taken respondedwell to IFNa therapy in vivo. Normal B-cells, CLL cells, Daudi-Burkitt's lymphoma cells, and hairy cells from patients resistant

to IFNa therapy did not exhibit these features.Effects upon the cytoskeleton are thought to be important

actions of IFN in many cell types and have been shown tocorrelate with cellular differentiation and subsequent responseto IFN (120). Ultrastructural studies of hematopoietic cellshave revealed that normal and malignant plasma cells characteristically possess "bubbling" or "blebbed" surface membranes

(121). Although IFNa-treated hairy cells do not appear tobecome morphologically identical to plasma cells, it is temptingto speculate that the appearance of areas of bubbling membraneon these cells represents progression towards plasmacytic differentiation. Previous studies demonstrating the synthesis ofnew organelles in cells from HCL patients responding to IFNatherapy lend support to this idea (122). The failure of thesecells to become morphologically indistinguishable from plasmacells is consistent with the hypothesis that IFNÂ«-induced hairycell differentiation becomes arrested and that late-acting factorsmay be needed for terminal differentiation to occur. However,in most patients, partial differentiation may be sufficient toeffectively control hairy cell proliferation.

Summary: Antiproliferative Effects of Interferon in Hairy CellLeukemia

We have demonstrated that ultrastructural changes occuronly in hairy cells displaying immunological modificationssuggestive of cellular differentiation and that these changescorrelate well with response to IFNa therapy. Affected cellsexhibit altered expression of several antigens that have beenimplicated as components required for normal B-cell activationand proliferation (33, 123, 124). The effects of such changesupon the proliferative capabilities of hairy cells might be crucialfor the favorable responses seen in HCL patients receivingIFNa to occur. In addition, the specificity of these immunological and morphological changes reflects both the unique sensitivity of hairy cells to IFNa and the inability of these cells toundergo the later stages of differentiation in a manner similarto that of other B-lymphocytes. The ability of IFNa to promotehairy cell maturation might explain the decreased growth factorresponsiveness exhibited by IFNa-treated hairy cells. Whetherthese changes are the result of oncogenic regulation by IFNahas not yet been determined. However, the failure of IFNÂ«toinduce terminal differentiation in hairy cells might indicate thelimits of IFNa as a single agent in HCL.

Our extensive review of the literature, laboratory investigations, and large clinical experience with HCL patients allowsus to draw several conclusions regarding the treatment of HCLand other malignancies with biological agents.

Mechanism of Interferon Action in Hairy Cell Leukemia.Hairy cell leukemia is a malignancy of partially activated B-

lymphocytes displaying a mature phenotype characterized bypositivity for HLA-DR, CD25, and tartrate-resistant acid phos-phatase. B-lymphocytes at this stage of differentiation engagein clonal proliferation in response to multiple cytokine growthfactors; correspondingly, hairy cell proliferation is uniquelydependent upon these same factors. B-cell growth is controlledin part by induction of differentiation, rendering cells lessresponsive to proliferative signals. Interferons have been demonstrated to produce such effects in normal and malignantlymphocytes (102, 104).

a-Interferon exerts direct effects upon the hairy cell, includ

ing modification of oncogene expression (107), induction ofspecific protein synthesis (125), and modulation of cytokineproduction (95). Present evidence indicates that these actionsfunction to induce partial hairy cell differentiation, renderingthe cell immune to growth factor stimulation, thereby inhibitingproliferation. The observed manifestations of this change include morphological alterations and a different immunophen-otype, features which have been clinically correlated with favorable responses to IFNa therapy (62, 87, 116). Hence, IFNaacts as a cytokine in reducing the hairy cell index, not as acytotoxic agent. Its actions upon the immune system have beenshown to facilitate the effective elimination of infectious pathogens, but not of the hairy cell.

The unique sensitivity of the hairy cell to IFNa, which is notexhibited by cells from CLL and related lymphoproliferativedisorders, can now be explained on the basis of their differingstages of maturation. B-cell growth factors are known to affectonly cells in particular phases of development. The mechanismsunderlying this differential sensitivity are not completely understood; however, they often depend upon activation/differentiation-dependent expression of growth factor receptors. CLL cellshave been shown to possess type I interferon receptors atdensities comparable to those of hairy cells (126), but theirfailure to express growth factor receptors susceptible to down-regulation by IFNa may indicate that these cells do not dependupon these factors for growth promotion. It is therefore notsurprising that IFNa, despite its ability to alter cellular processes in leukemic cells from CLL, does not affect cellular growthto any great extent in CLL. This concept is supported by severalstudies demonstrating that hairy cells taken from patients resistant to IFNa therapy display atypical immunophenotypes,placing them in differentiation stages of greater or lesser maturity than that of the typical hairy cell. Interestingly, cells fromneither CLL nor the two major recognized variant (resistant)HCL types express the IL-2R (127, 128). This finding probably

indicates that the response to IFNa therapy depends uponinterruption of IL-2-sustained hairy cell proliferation or that

the presence of CD25 marks a hairy cell differentiation stagein which receptors for other essential growth factors are expressed and down-regulated by IFNa.

Implications for the Biotherapy of Hairy Cell Leukemia andOther Malignancies. Clinical trials using IFNa against a widevariety of malignancies have shown that favorable responses totherapy are only infrequently seen when IFNa is used in high

1063




11,

12.

13.

15.

16.

17.

19.

doses as a chemotherapeutic agent rather than as a cytokine.The evidence presented in this review supports the idea that thebeneficial effects of IFNa in cancer are more likely to be due tothe role of IFNa as a cytokine rather than a cytotoxic agent. gHowever, because IFNa acts as an antiproliferative biologicalresponse modifier in HCL, complete bone marrow remissionsare quite uncommon. IFNa has not been shown to causeirreversible damage to hairy cell progenitors; as a result, nearly 10.all HCL patients eventually relapse after termination of IFNatherapy (21). In addition, some patients initially responding toIFNa develop resistance to its effects; these cases may involvehairy cells that partially differentiate in response to IFNa butfor some reason continue to proliferate (62). In vitro studieshave clearly demonstrated that IFNa is not capable of inducingterminal differentiation in hairy cells. The high relapse ratesseen in HCL and the development of acquired resistance in 14some HCL cases indicate that the inability of IFNa to irreversibly harm hairy cell progenitors and its failure to induce terminal differentiation may define the limits of IFNa therapy inHCL.

Although these limitations may apply when IFNa is used asa lone agent, the possibility of overcoming resistance to IFNausing logical combined biotherapy has not yet been assessed.The recent demonstration (129) that IFNa induces receptorsfor melanocyte-stimulating hormone on melanoma cells, caus- 18ing them to differentiate in response to melanocyte-stimulatinghormone, displays the rationale behind this approach and illustrates the true potential of cancer biotherapy. IFNa-resistanthairy cells have been demonstrated to possess greater or lesserdegrees of maturity in comparison to typical hairy cells; we andothers (127, 128) have suggested that this difference may formthe basis of the resistant phenotype. It is therefore reasonable 20to speculate that the addition of early-acting cytokines targetedto the appropriate level of hairy cell maturity might cause lessmature hairy cells to become sensitive to IFNa, while agents 21associated with late events in B-cell development might inducefurther differentiation in IFNa-treated resistant cells. Ourgrowing understanding of the nuclear and cytosolic events thatcharacterize B lymphocyte differentiation makes HCL an excellent model in which to examine logical combined cancerbiotherapy, the "fourth arm" of modern cancer treatment ( 130). 24

In addition, the demonstrated effects of IFNa upon the lym-phocytic cell cycle suggest that further elucidation of thesepathways may eventually allow us to temporally coordinatechemotherapy with biotherapy. Following multidisciplinarystudies of oncogene expression, growth factor responsiveness, 26and tumor cell differentiation, these approaches might be rationally and effectively utilized in other malignancies as well.

27.

References

1. Baron, S., Tyring, S. K., Fleischmann, W. R., Coppenhauer, D. H., Niesei. 28.D. W., Kumpel, G. R., Stanton, G. J., Hughes, T. K. The interferons:mechanisms of action and clinical applications. JAMA, 266: 1375-1383,1991. 29.

2. Creagen, E. T. Regional and systemic strategies for malignant melanoma.Mayo Clin. Proc., 64: 852-860, 1989.

3. Glashan R. W. A randomized controlled study of intravesical a-2b-inter-feron in carcinoma in situ of the bladder. J. Urol., 144: 658-661, 1990. 30.

4. Merimsky, O., Inhar, M., Merimsky, F., Koyner, F., Spitzer, F., Laufer, R.,Braf, Z., and Chaitchik, S. Phase II study of recombinant interferon alpha-c in patients with metastatic renal cell carcinoma. Mol. Biother., 2: 155- 31159, 1990.

5. Alexanian, R.. and Barlogie. B. New treatment strategies for multiplemyeloma. Am. J. Hematol.. 35: 194-198, 1990.

6. Foon, K. A., and Bunn, P. A. a-Interferon treatment of cutaneous 32T cell lymphoma and chronic lymphocytic leukemia. Semin. Oncol.,13 (Suppl. 5): 35-39, 1986.

1064

22.

23.

25.

Alimena, G., MorrÃ , E., Lazzarino, M., Liberati, A. M., Montefusco, E.,Inverardi, D., Bernasconi, P., and Mancini, M., Donti, E., Grignani, F.,Bernasconi, C., Dianzani, F., and Mandelli, F. Interfereon a-2b as therapyfor Ph'-positive chronic myelogenous leukemia: a study of 82 patientstreated with intermittent or daily administration. Blood, 72:642-647,1988.Talpaz, M., Kurzrock, R., Kantarjian, II. and Gutterman, J. Therapy ofchronic myelogenous leukemia with interferons. Cancer Surv., 8: 793-798,1989.Colomb, H. M., Ratain, M. J., and Moormeier, J. What is the choice oftreatment for hairy cell leukemia? J. Clin. Oncol., 7: 156-158, 1989.Mintz, U., and Golomb, H. M. Splenectomy as initial therapy in twenty-six patients with leukemic reticuloendotheliosis (hairy cell leukemia). CancerRes., 39: 2366-2370, 1979.Yam, L. T., Chaudhry, A. A., and Janckila, A. J. Impaired granulocytereserve and leukocyte mobilization in leukemic reticuloendotheliosis. Ann.Intern. Med., 87: 444-446, 1977.Yam, L. T., Li, C. Y., and Lam, K. W. Tartrate-resistant acid phosphataseisoenzyme in the reticulum cells of leukemic reticuloendotheliosis. N. Engl.J. Med., 284: 357-360, 1971.Polliack, A., Braylan, A., and Golomb, H. M. Scanning electron microscopyof hairy cells in hairy cell leukemia. Lancet, 2: 1013-1014, 1974.Fu, S. M., Winchester, R. J., Rai, K. R., and Kunkel, H. C. Hairy cellleukemia: proliferation of a cell with phagocytic and B lymphocyte properties. Scand. J. Inumi noi., 3: 847-851, 1974.Cawley, J. C., Burns, G. F., Nash, T. A., Higgy, K. E., Child, J. A., andRoberts, B. E. Hairy cell leukemia with T-cell features. Blood, 51: 61-69,1978.Burns, G. E., Cawley, J. C., Worman, C. P., Karpas, A., Barker, C. R.,Goldstone, A. 11.. Hayhoe, F. G. H. Multiple heavy chain isotypes on thesurface of the cells of hairy cell leukemia. Blood, 52: 1132-1147, 1978.Jansen, J., LeBien, T. W., and Kersey, J. The phenotype of neoplastic cellsof hairy cell leukemia studied with monoclonal antibodies. Blood, 59: 609-614, 1982.Korsemeyer, S. J., Greene, W. C., Cossman, J., Hsu, S., Jensen, J. P.,Neckers, L. M., Marshall, S. L., Bakhshi, A., Depper, J. M., Leonard, W.J., Jaffe, E. S., and Waldmann, T. A. Rearrangement and expression ofimmunoglobulin genes and expression of Tac antigen in hairy cell leukemia.Proc. Nati. Acad. Sci. USA, 80:4522-4526, 1983.Caligaris-Cappio, F., Pizzolo, G., Chilosi, M., Bergui, L., Semenzaio, G.,Tesio, L., Morittu, L., Malavasi, F., Gobbi, M., Schwarting, R., Campana,D., and Janossy, Y. Phorbol ester induces abnormal chronic lymphocyticleukemia cells to express features of hairy cell leukemia. Blood, 66: 1035-1042, 1985.Gaziti, Y., Gurfel, D., Leizerowitz, R., Feinstein, H., and Polliack, A.Heterogeneity in TPA-induced differentiation of B-chronic lymphocyticleukemia cells: development of hairy cell or plasmacytoid features are timeand dose dependent. Cancer Invest., 5: 309-319, 1987.Doane, L. L., Ratain, M. J., and Golomb, H. M. Hairy cell leukemia:current management. Hematol. Oncol. Clin. North Am., 4:489-502, 1990.Ratain, M. J., Vardiman, J. W., and Golomb, H. M. The role of interferonin the treatment of hairy cell leukemia. Semin. Oncol., li (Suppl 2): 21-28,1986.Ratain, M. J., Vardiman, J. W., Barker, C. M., and Golomb, H. M.Prognostic variables in hairy cell leukemia after splenectomy as initialtherapy. Cancer (Phila.), 62: 2420-2424, 1988.Johnston, J. B., Eisenhauer, E., Corbett, W. E., Scott, J. G., and Zaentz,S. D. Efficacy of 2'-deoxycoformycin in hairy cell leukemia: a study of the

National Cancer Institute of Canada Clinical Trials Group. J. Nati. CancerInst., Â«0:765-769, 1988.Kraut, E. H., Bouroncle, B. A., and Grever, M. R. Pentostatin in thetreatment of advanced hairy cell leukemia. J. Clin. Oncol., 7: 168-172,1989.Spiers, A. S., Moore, D., Cassileth, P. A., Harrington, D. P., Cummings,F. J., Neiman, R. S., Bennett, J. M., and O'Connell, M. J. Remissions inhairy cell leukemia with pentostatin (2'-deoxycoformycin). N. Eng). J. Med.,316: 825-830, 1987.Urba, W. J., Baseler, M. W., Koop, W. C., Sties, R. G., Clark, J. W., Smith,J. W. II, Coggin, D. L., and Longo, D. L. Deoxycoformycin inducedimmunosuppression in patients with hairy cell leukemia. Blood, 73: 38-46,1989.Piro, L. D., Carrera, C. J., Carson, D. A., and Beutler, E. Lasting remissionsin hairy-cell leukemia induced by a single infusion of 2-chlorodeoxyadeno-sine. N. Engl. J. Med., 322: 1117-1121, 1990.Ratain, M. J., Golomb, H. M., Vardiman, J. W., Westbrook, C. A., Barker,C., Hooberrman, A., Bitter, M. A., and Daly, K. Relapse after interferonalfa-2b therapy for hairy cell leukemia: analysis of prognostic variables. J.Clin. Oncol., 6: 1714-1721, 1988.Seshardi, R. S., Brown, E. J., and Zipursky, A. Leukemic reticuloendotheliosisâ€”a failure of monocyte production. N. Engl. J. Med., 295: 181-184,1976.Ruco, L. R., Procopio, A., Maccalini, V., Calogero, A., I nini. S., Annino,I., Mandelli, E., and Baroni, C. D. Severe deficiency of natural killer activityin the peripheral blood of patients with hairy cell leukemia. Blood, 61:1132-1137,1983.Mackowiak, P. A., Demian, S. E., Sutker, W. L., Murphy, F. K., Smith, J.W., Tompsett, R., Sheehan, W. W., and Luby, J. P. Infections in hairy cellleukemia. Clinical evidence of a pronounced defect in cell-mediated immu-




nity. Am. J. Med., 6Â«:718-724, 1980.33. Tsudo, M., Uchiyama, T., and Uchino, H. Expression of Tac antigen on

activated normal human B cells. J. Exp. Med., 160: 612-617, 1984.34. Rubin, L. A., Rumimi. C. C., Fritz, M. E., Biddison, W. E., Boutin, B.,

Yarchoan, R., and Nelson, D. L. Soluble interleukin 2 receptors are releasedfrom activated human lymphoid cells in vitro. J. Immunol., 135: 3172-3177, 1985.

35. Chilosi, M., Semenzaio, G., Getto, G., Ambrosetti, A., Fiore-Donati, L.,Perona, G., Berton, G., Lestani, M., Scarpa, A., Agostini, C., Trentin, L.,Zambello, R., Masciarelli, M., Dazzi, F., Vinante, F., Caligaris-Cappio, F.,and Pizzolo, G. Soluble interleukin-2 receptors in the sera of patients withhairy cell leukemia: relationship with the effect of recombinant a-interferontherapy on clinical parameters and natural killer in vitro activity. Blood, 70:1530-1535, 1987.

36. Steis, R. G., Marcon, L., Clark, J., Urba, W., Longo, D. L., Nelson, D. L.,and Maluish, A. E. Serum soluble IL-2 receptor as a tumor marker inpatients with hairy cell leukemia. Blood, 71: 1304-1309. 1988.

37. Richards, J. M, Mick, R., Latta, J. M., Daly, K., Ratain, M. J., Vardiman,J. W., and Golomb, H. M. Serum soluble interleukin-2 receptor is associatedwith clinical and pathologic disease status in hairy cell leukemia. Blood, 76:1941-1945, 1990.

38. Olofsson, T., and Olsson, I. Suppression of normal granulopoiesis in vitroby a leukemia-associated inhibitor (LAI) of acute and chronic leukemia.Blood, 55: 975-982, 1980.

39. Broxmeyer, H. E., Jacobsen, N., Kurland. J., Mendelsohn, N., and Moore,M. A. S. In vitro suppression of normal granulocytic stem cells by inhibitoryactivity derived from human leukemia cells. J. Nati. Cancer Inst., 60: 497-511, 1978.

40. Richman, C. M., and Golomb, H. M. Hairy cell leukemia: effect of'hairy'cells on normal granulopoiesis in vitro. Exp. Hematol., 7: 411-415, 1979.

41. Taniguchi, N., Kuratsune, H.. Kanamaru, A., Tokumine. V. Tagawa, S.,Machii, T., and Kitani, T. Inhibition against CFU-C and CFU-E colonyformation by soluble factor(s) derived from hairy cells. Blood, 73:907-913,1989.

42. Lauria, F., Bagnara, G. P., Caiani, L., Gaggiolo, L., Guarini, A., Raspadori.I>., Foa, R., Bellone, G., Buzzi, M., Gugliotta, L., Zauli, G., Giovannini,M., /in/ani. P. L., and Tura, S. The inhibitory effect of serum from hairy-cell leukaemia patients on normal progenitor cells may disappear followingprolonged treatment with n-interferon. Br. J. Haematol., 72:497-501.1989.

43. Lindemann, A., Ludwig, W. D., Oster, W., Mertelsmann, R., and Herrmann, F. High-level secretion of tumor necrosis factor-a contributes tohematopoietic failure in hairy cell leukemia. Blood. 73: 880-884, 1989.

44. Lambertenghi-Deliliers, G., Benazzi, E., Cortelezzi, A., and Polli, E. E.Bone marrow histological modifications induced by a interferon in hairycell leukemia. Acta Haematol. Basel, 77: 203-207, 1987.

45. Quesada, J. R., Hersh, E. M., Manning, J., Reuben, J., Keating, M.,Schnipper, E., Uri, L., and Gutterman, J. U. Treatment of hairy cellleukemia with recombinant n-interferon. Blood, 68: 493-497, 1986.

46. Perussia, B., Kobayashi, M., Rossi, M. E., Anegon, I., and Trinchieri, G.Immune interferon enhances functional properties of human granulocytes.Role of Fc receptors and effect of lymphotoxin tumor necrosis factor andgranulocyte-macrophage colony stimulating factor. J. Immunol., Â¡38:765-774, 1987.

47. Hofmann, V., Fehr, J., Sauter, C., and Oliino, J. Hairy cell leukemia: aninterferon deficient disease? Cancer Treat. Rev., 12 (Suppl B): 33-37. 1985.

48. Yoshioka, H., Shimizu, N., Maeta, M.. Koga. S., Hashimoto. K., andNakamura, K. Immunological evaluation after splenectomy for hairy cellleukemia: a case report. Jpn. J. Surg., 17: 287-301, 1987.

49. Ho, A. D., Grossman, M., Trumper, L., Pezzullo, A., Fraile, H.. Boedewadl-Raozun, S., Papendick, U., God, B., Zwingers, T., and Hunstein, W. Clinicalimplications of increased plasma levels of CD8 in patients with hairy cellleukemia. Blood, 75: 1119-1124, 1990.

50. Worman, C. P., Nethersell, A. B., Bottomley, J. M., Apostolov, K., Barker,W., and Cawley, J. C. Natural IFN-Â« therapy in hairy-cell leukaemia(Namalva-type IFNâ€”Wellferon). Klin. Wochenschr., 65: 685-687, 1987.

51. Schwanz, R. H. The role of gene products of the major histocompatibilitycomplex in T cell activation and cellular interactions. In: W. E. Paul (ed.).Fundamental Immunology, pp. 379-438. New York: Raven Press, 1984.

52. Carter, R. H., Drebin, J. A., Schallen, S., Perry, L., and Green, M. I.Regulation of the immune response to tumor antigens. IX. In vitro Lyt.1+2â€”cell proliferative responses to cellbound or subcellular tumor antigen.J. Immunol., 130: 997-1003, 1983.

53. Forni, G., Shevach, E. M., and Green, I. Mutant lines of guinea pig L2Cleukemia. I. Deletion of la alloantigens is associated with a loss of imiuunogenicity of tumor-associated antigens. J. Exp. Med., 143: 1067-1081,1975.

54. Hokland, M., Heron, I., Hokland, P., Basse, P., and Berg, K. Measurementsof changes in histocompatability anligens induced by inlerferons. MelhodsEnzymol., 119: 688-693, 1986.

55. Carson, D., Dokhelar, M., Wakasugi. H., Mishal, Z., and Tursz, T. HLAClass-I and Class-II anligen expression by human leukemic K562 cells andby Burkitt-K562 hybrids: modulation by differentiation inducers and interferon. Exp. Hematol., 13: 885-890, 1985.

56. Giancomini, P., Fisher, P. B., Duigou, G. J., Gambari, R., and Natali. P.G. Regulalion of class II MHC expression by inlerferons: insighls into themechanism of action of interferon (review). Anticancer Res.. *: 1153-1161,1988.

57. Baldini, L., Cortelezzi. A.. Polli, N., Neri, A., Nobili, L., Maiolo, A. T.,Lambertenghi-Deliliers, G.. and Polli, E. E. Human recombinant interferona-2C enhances Ihe expression of class II HLA anligens on hairy cells. Blood,67:458-464, 1986.

58. Huang, A. T., and Mold, N. G. Changes in hairy cells afler alpha-inlerferontrealmenl as measured by flow cytometry. Leukemia (Ballitnore), 1: 308-311, 1987.

59. Gamliel, H., Gurfel, D., Wu, S., and Golomb, H. M. Inlerferon-inducedsurface alleralions in hairy cells. A review. Scanning Microsc., 2:485-492,1988.

60. Gressler, V. H., Weinkauff, R. E., Franklin, W. A., and Golomb, H. M.Modulation of Ihe expression of major hislocompatibility antigens onsplenic hairy cells-differential effecl upon in vitro ireatment with a-2b-interferon, 7-interferon, and inlerleukin-2. Blood, 72: 1048-1053, 1988.

61. Griffilhs, S. D., and Cawley, J. C. The beneficial effecls of a-inlerferon inhairy cell leukaemia are noi due lo NK medialed cyloloxicity. Leukemia(Baltimore), /: 372-376, 1987.

62. Till, K. J., and Cawley, J. C. Phenotypic changes on hairy cells exposed invitro lo inlerferons: a quanlitative FACS sludy. Br. J. Haemalol., 72: 378-386, 1989.

63. Silver, H. K. B., Salinas, F. A., and Kong, S. Role of interferons in immuneregulation. In: H. K. B. Silver (ed.), Interferons in Cancer Treatment, pp.1-9. Toronto, Ontario, Canada: MES Medical Education Services, 1986.

64. Casti, G., Aulitzky, W., Leiler, E., Flener, R., and Huber, C. a-Inlerferoninduces remission in hairy cell leukemia wilhoul enhancement of naturalkilling. Blut, 52: 273-279, 1986.

65. Foa, R., Lauria, F., Lusso, P., Raspadori, D., Fierro, M. T., Tazzari, P. L.,Caudana, L., and Matera, L. Phenolypic and funclional characlerizalion ofIhe circulaling NK comparlmenl in hairy cell leukaemia. Clin. Exp. Immunol., 64: 392-398, 1986.

66. Oshimi, K., Oshimi, Y., Moloji, T., Kobayashi, S., and Mizoguchi, H. Lysisof leukemia and lymphoma cells by aulologous and allogeneic interferon-activated blood mononuclear cells. Blood, 61: 790-798, 1983.

67. Lee, S. H., Kelly, S., ChiÃ¹, H., and Slebbing, N. Slimulalion of naturalkiller cell activily and inhibilion of proliferalion of various leukemic cellsby purified human leukocyle inlerferon subtypes. Cancer Res., 42: 1312-1316, 1982.

68. Yokes, E. E., Bitter, M. A., Ratain, M. J., Prystowsky, M. B., Daly, K., andGolomb, H. M. Flow cytometry in hairy cell leukemia before and duringinterferon alfa-2b Iherapy. Cancer (Phila.), 59: 1987-1991, 1987.

69. Foa, R., Fierro, M. T., Lusso, P., Bonferroni, M., Raspadori, D., Buzzi,M., Zinzani, P. L., Resegotli, L., and Lauria, F. Effecl of a-interferon onihe immune syslem of palienls wiih hairy cell leukemia. Leukemia (Baltimore),/: 377-379, 1987.

70. Trenlin, L., Zambello, R., Agoslini, C., AmbroseÂ«!, A., Chisesi, T., Rai-mondi, R., Bulian, P., Pizzolo, G., and Semenzaio, G. Mechanisms accounl-ing for the defeclive nalural killer activity in patients with hairy cell leukemia. Blood, 75: 1525-1530, 1990.

71. Semenzaio, G., Pizzolo, G., Agoslini, C., Ambroselli, A., Zambello, R.,Trentin, L., Luca, M., Masciarelli, M., Chilosi, M., Vinante, F., Perona,G., and Cello, G. a-inlerferon aclivales the natural killer system in patienlswith hairy cell leukemia. Blood, 68: 293-296, 1986.

72. Foon, K. A., Maluish, A. E., Abrams, P. G., Wrighlinglon, S., Stevenson,H. C., Alarif, A., Fer, M. F., Overton, W. R., Poole, M., Schnipper, E. F.,Jaffe, E. S., and Herberman, H. B. Recombinanl leukocyte A interferontherapy for advanced hairy cell leukemia. Therapeutic and immunologieresults. Am. J. Med., 80: 351-356, 1986.

73. Sigaux, F., Castaigne, S., Lehn, P., Dupuy, E., Billard, C., Gluckman, J.,Boiron, M., Falcoff, E., Flandrin, G., and Degos, L. a-inlerferon in hairycell leukaemia: direcl effects on hairy cells or indirecl cyloloxicily? Ini. J.CancerS1!//)/)/. /: 2-8, 1987.

74. Trenlin, L., Ambrosetti, A., Zambello, R., Agostini, C., Masciarelli, M.,Vinante, F, Chilosi, M., Pizzolo, G., Cello, G., and Semenzaio, G. a-Inlerferon aclivaled cylotoxic lymphocyles in hairy cell leukemia palienls:evalualion of cytotoxic events. Leuk. Res., //: 843-847, 1987.

75. Starling, G. C., Nimmo, J. C., and Hart, D. N. Hairy cell leukemia cellsare relatively NK-insensitive targets. Pathology, 20: 361-365, 1988.

76. Balmer, C. M. The new a interferons. Drug Intell. Clin. Pharmacol., 19:887-893, 1985.

77. Roth, M. S., and Foon, K. A. a Interferon in the treatment of hÃ©matologiemalignancies. Am. J. Med., */: 871-882, 1986.

78. Revel, M. The interferon system in man: nature of the interferon moleculesand mode of action. In Y. Becker (ed.), Antiviral Drugs and Interferon, pp.357-433. Boslon: Martinus Nijhoff Publishing, 1984.

79. Freidman, R. M. Modulalion of growih and differentiation by interferons.In: H. K. B. Silver (ed.), Inlerferons in Cancer Treatment, pp. 11-16.Toronto, Onlario, Canada: MES Medical Educalion Services, 1986.

80. Ford, R. J., Yoshimura, L., Morgan, J., Quesada, J., Monlagna, R., andMaizel, A. Growih factor-medialed tumor cell proliferalion in hairy cellleukemia. J. Exp. Med., 162: 1093-1098, 1985.

81. Ford, R. J., Kwok, D., Quesada, J., and Sahasrabuddhe, C. G. Produclionof B cell growth faclor(s) by neoplaslic B cells from hairy cell leukemiapalienls. Blood, 67: 573-577, 1986.

82. Paganelli, K. A., Evans, S. S., Han, T., and Ozer, H. B cell growih factor-induced proliferation of hairy cell lymphocytes and inhibition by lype Iinlerferon in vitro. Blood, 67: 937-942, 1986.

83. Genol, E., Billard, C., Sigaux, F., Malhiol, C., Degos, L., Falcoff, E., and

1065




Kolb, J. P. Proliferative response of hairy cells to B cell growth factor 107.(BCGF): in vivo inhibition by interferon-a and in vitro effects of inlerferon-a, -ÃŸ,and -y. Leukemia (Baltimore). /: 590-596, 1987.

84. Mignari, M. C., Cerosa, F., Carra, G., Accolla, R. S., Moretta, Z., Zubler, 108.R. H., Waldmann, T. A., and Moretta, L. Human interleukin-2 promotesproliferation of activated B cells via surface receptors similar to those ofactivated T cells. Nature (Lond.), 312:641-643, 1984.

85. Ralph, P., Jeong, G., Weite, K., Mertelsmann, R., Rabin, H., Henderson, 109.L. E., Souza, L. M., Boone, T. C., and Robb, R. J. Stimulation of immu-noglobulin secretion in human B lymphocytes as a direct effect of highconcentrations of IL-2. J. Immunol., 133: 2442-2445, 1984. 110.

86. Gordon, J., Webb, A. J., Walker, L., Guy, G. R., and Rowe, M. Evidencefor an association between CD23 and the receptor for a low molecularweight B cell growth factor. Eur. J. Immunol., 16: 1627-1630, 1986. 111.

87. Gamliel, H., Brownstein. B. H., Gurfel, D., Wu, S., Rosner, M. C., andGolomb, H. M. B-cell growth factor-induced and a-interferon-inhibitedproliferation of hairy cells coincides with modulation of cell surface antigens. 112.Cancer Res.. 50: 4111-4120, 1990.

88. Kelker, H. C, Oppenheim, J. D., Stone-Wolff, D., Henriksen-Destefano,li.. Aggarwal, B. B., Stevenson, H. C., and Vilcek, J. Characterization ofhuman tumor necrosis factor produced by peripheral blood monocytes and 113.its separation from lymphotoxin. Int. J. Cancer, 36:69-73, 1985.

89. Carswell, E. A., Old, L. J., Kassel, R. L., Green, S., Fiore, N., andWilliamson, B. An endotoxin-induced serum factor that causes necrosis of 114.tumors. Proc. Nati. Acad. Sci. USA, 72: 3666-3670, 1975.

90. Cordingley, F. T., Bianchi, A., Hoffbrand, A. V.. Reittie, J. E., Heslop, H.E., Vyakarnam, A., Turner, M., Meager, A., and Brenner, M. K. Tumournecrosis factor as an autocrine tumor growth factor for chronic B-cell 115.malignancies. Lancet, /: 969-971, 1988.

91. Kehrl, J. H., Miller, A., and Fauci, A. S. Effect of TNF-a on mitogenactivated human B cells. J. Exp. Med., 166: 786-791, 1987. 116.

92. Billard, C., Sigaux, F., and Wietzerbin, J. IFN-a in vivo enhances tumornecrosis factor receptor levels on hairy cells. J. Immunol., 145: 1713-1718,1990. 117.

93. Barut, B. A., Cochran, M. K., O'Hara, C, and Anderson, K. C. Responsepatterns of hairy cell leukemia to B-cell mitogens and growth factors. Blood,76:2091-2097,1990. 118.

94. Foa, R., Massaia, M., Cardona, S., Tos, A. G., Bianchi, A., Attisano, C.,Guarini, A., di Celle, P. F., and Fierro, M. T. Production of tumor necrosisfactor-Â«by B-cell chronic lymphocytic leukemia cells: a possible regulatoryrole of TNF in the progression of the disease. Blood, 76: 393-400. 1990. 119.

95. Heslop, H. E., Bianchi, A. C., Cordingley, F. T., Turner, M., Chandima.W., de Mel, C. P., Hoffbrand, A. V., and Brenner, M. K. Effects ofinterferon-a on autocrine growth factor loops in B lymphoproliferativedisorders. J. Exp. Med., /72: 1729-1734, 1990. 120.

96. Michalevitz, R., Lifshitz, D., and Revel, M. Interferon beta 2/interleukin-6and interleukin-3 synergize in stimulating proliferation of human early 121.hematopoietic progenitor cells. Scanning Microsc., 3:1143-1150, 1989.

97. Greenberg, M. E., and Ziff, E. B. Stimulation of 3T3 cells induces transcription of the c-/0Ã¬protooncogene. Nature (Lond.), 311: 433-438, 1984.

98. Grausz, J. D., Fradelizi, D.. Dautry, F., Monier, R., and Lehn, P. Modu- 122.lation of i- Im and c-myc mRNA levels in normal human lymphocytes bycalcium ionophore A23187 and phorbol ester. Eur. J. Immunol., 16: 1217-1221. 1986. 123.

99. Westin, F. H., Wong-Staal. F., Gelman, E. P., Dalla-Pavera, R., Papas, T.S., Lautenberger, J. A., Eva. A., Redy, E. P., Tronick, S. R., Aaronson, S.A., and Gallo. R. C. Expression of cellular homologues of retroviral onegenes in human hematopoietic cells. Proc. Nati. Acad. Sci. USA, 79: 2490- 124.2494, 1982.

100. Einat, M., Resnitzky, D., and Kimchi, A. Close link between reduction ofc-myc expression by interferon and G0/G, arrest. Nature (Lond.), 313: 597- 125.600, 1985.

101. Roos, G., Leanderson, T., and Lundgren, E. Interferon-induced cell cyclechanges in human hematopoietic cell lines and fresh leukemic cells. Cancer 126.Res., â€¢¿�Â«.â€¢2358-2362,1984.

102. Resnitzky, D., Yarden, A., Zipori, D., and Kimchi, A. Aulocrine-/3-relatedinterferon controls c-myc suppression and growth arrest during hematopoietic cell differentiation. Cell, 46: 31-40, 1986. 127.

103. Jonak, G. J., and Knight, E., Jr. Selective reduction of c-myc mRNA inDaudi cells by human /3-interferon. Proc. Nati. Acad. Sci. USA, */: 1747-1750, 1984.

104. Dani, C. H., Mechti, N., Piechaczyk, M., Lebleu. B., Jeanteur, P. H., and 128.Blanchard, J. M. Increased rate of degradation of c-myc mRNA in inter-feron-treated Daudi cells. Proc. Nati. Acad. Sci. USA, 82:4896-4899,1985.

105. Reitsma, P. H., Rothberg, P. G., Astrin, S. M., Trial, J., Bar-Shavit, Z.,Hall, H., Teitelbaum, S. L., and Kahn, A. J. Regulation of myc gene 129.expression in III. 60 leukaemia cells by a vitamin D metabolite. Nature(Lond.), 306: 492-449, 1983.

106. Grosso, L. E., and Pilot, H. C. Modulation of c-myc expression in the HL- 130.60 cell line. Biochem. Biophys. Res. Commun.. 119:473-480, 1984.

Lehn, P, Sigaux, F., Grausz, D., Loiseau, P., Castaigne, S., Degos, L.,Flandrin, G., and Dautry, F. c-myc and c-fos expression during interferon-a therapy for hairy cell leukemia. Blood, 68: 967-970, 1986.Exley, R., Gordon, J., and Clemens, M. J. Induction of B-cell differentiationantigens in interferon- or phorbol-ester treated Daudi cells is impaired byinhibitors of ADP-ribosyltransferase. Proc. Nati. Acad. Sci. USA, 84:6467-6470, 1987.Kimchi, A. Increased levels of interferon-induced (2'-5') oligoadenylatesynthetase in mature T-lymphocytes and in differentiated Friend erythroleu-kemic cells. J. Interferon Res., /: 559-569, 1981.Ostlund, L., Einhorn, S., Robert, K. H., Juliusson, G., and Biberfeld, B.Chronic B-lymphocytic leukemia cells proliferate and differentiate followingexposure to interferon in vitro. Blood, 67: 152-159, 1986.Anderson, K. C., Boyd, A. W., Fisher, D. C., Leslie, D., Schlossman, S. F.,and Nadler, L. M. Hairy cell leukemia: a tumor of pro-plasma cells. Blood,65:620-629, 1985.Polliack, A., Gurfel, D., Leizerowitz, R., Feinstein, H., and Gazitt, Y.Phorbol ester (TPA)-induced surface membrane alterations in B-type hairycells and lymphocytic leukemia cells. Scanning Electron Microsc., I'/. 4:1523-1535, 1986.Skinnider, L., Catovsky, D., and Incarbone, E. The effect of tetradecanoyl-12,13-phorbol acetate on hairy cells. Scand. J. Haematol., 33: 301-308,1984.Lockney, M. W., Golomb, H. M., and Dawson, G. Unique cell surfaceglycoprotein expression in hairy cell leukemia: effect of phorbol ester tumorpromoters on surface glycoproteins, cell morphology, and adherence. CancerRes., 42: 3724-3728, 1982.Lockney, M. W., Dawson, G., and Golomb, H. M. Effect of phorbol estertumor promoters on hairy cell leukemic cells. Semin. Oncol., 11: 433-438,1984.Gressler, V. H., Weinkauff, R. E., Franklin, W. A., and Golomb, H. M. Isthere a direct differentiation-inducing effect of human recombinant interferon on hairy cell leukemia in vitral Cancer (Phila.), 64: 374-378, 1989.Schlossman, S. F., Chess, L., Humphreys, R. E., and Strominger, J. L.Distribution of la like molecules on the surface of normal and leukemiccells. Proc. Nati. Acad. Sci. USA, 73: 1288-1293, 1976.Polliack, A., Leizerowitz, R., and Gazitt, Y. Phorbol esters and hairy cellleukemia: effects on cell morphology and surface membrane features andcomparison with other B cell leukemias. Leukemia (Baltimore), /: 311-315,1987.Halper, J. P., Fu, S. M., Gottlieb, A. B., Winchester, R. J., and Kunkel, H.G. Poor mixed lymphocyte reaction stimulatory capacity of leukemic B cellsof chronic lymphocytic leukemia patients despite the presence of la antigens.J. Clin. Invest., 64: 1141-1147, 1979.Boiron, M. Future developments in interferon therapy. Int. J. Cancer, Suppl.A-41-44, 1987.

Polliack, A., Leiserowitz, R., Korkesh, A., Matzner, Y., Prokocimer, M.,and Gamliel, H. Plasma cell leukemia and myeloma. A scanning electron-microscopic study of cell surface features in six cases. Am. J. Clin. I'athol..75:834-838, 1981.Samuels, B. L., Rosner, M. C., Giomelti, C. S., Golomb, H. M., andBrownstein, B. H. Action of interferons in hairy cell leukemia. Leukemia(Baltimore), /: 365-369, 1987.Pezzullo, A., Dorken, B., Rabinovitch, P. S., Ledbetter, J. A., Moldenhauer,G., and Clark, E. A. CD19 monoclonal antibody HD37 inhibits anti-immunoglobulin-induced B cell activation and proliferation. J. Immunol.,138: 2793-2799, 1987.Pezzullo, A., Dorken, B., Moldenhauer, G., and Clark, E. A. Amplificationof human B-cell activation by a monoclonal antibody to the !*â€¢cell-specificantigen CD22, Bp 130/40. J. Immunol., 138: 98-103, 1987.Samuels, B. L., Golomb, H. M., and Brownstein, B. H. Different proteinsare induced by a- and -y-interferon in hairy cell leukemia. J. Biol. ResponseModif., 6: 268-274, 1987.Faltynek, C. R., Princier, G. L., Rossio, J. L., Ruscelli, F. W., Maluish, A.E., Abrams, P. G., and Foon, K. A. Relationship of ihe clinical responseand binding of recombinant interferon a in patients wilh lymphoproliferalivediseases. Blood, 67: 1077-1082, 1986.Sainati, L., Matutes, E., Mulligan, S., de Oliveira, M. P., Rani, S., Lampen,I. A., and Catovsky, D. A varianl form of hairy cell leukemia resistanl to a-intor toriin: clinical and phenotypic characteristics of 17 patients. Blood, 76:157-162,1990.Lauria, F., Raspador!, B. S., Foa, R., Zinzani, P. L., Buzzi, M., Gugliotla,M. D., Macchi, S., and Tura, S. Reduced hÃ©matologieresponse to a-interferon therapy in patients with hairy cell leukemia showing a peculiarimmunologie phenotype. Cancer (Phila.), 65: 2233-2236, 1990.Kameyama, K., Tanaka, S., Ishida, Y., and Hearing, V. J. Interferonsmodulate Ihe expression of hormone receptors on the surface of murinemelanoma cells. J. Clin. Invest., S3: 213-221, 1989.Goldstein, D., and Laszlo, J. The role of inlerferon in cancer therapy: acurrenl perspeclive. CA Cancer J. Clin., 38: 258-277, 1988.

1066



1992;52:1056-1066. Cancer Res Suresh Vedantham, Haim Gamliel and Harvey M. Golomb of Effective Cancer BiotherapyMechanism of Interferon Action in Hairy Cell Leukemia: A Model

Updated version

http://cancerres.aacrjournals.org/content/52/5/1056

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/52/5/1056To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



mechanism of interferon action in hairy cell leukemia: a model … · hairy cell leukemia is a...

Documents