The occurrence of thrombotic thrombocytopenic purpura (TTP) in cancer patients receiving chemotherapy has been well established; although this entity is rare, its clinical importance seems to be growing. We describe 3 cases of TTP developing in cancer patients receiving different chemotherapeutic regimens. Using a sensitive high-performance liquid chromatographic method, we evaluated the stable nitric oxide end products, nitrite and nitrate, in the plasma of these patients. Nitric oxide is one of the key components involved in maintaining the normal nonthrombogenicity of the vascular endothelium. In our 3 patients, we found increased nitrate titers that were substantially higher than those observed in patients with de novo TTP. The observed increased release of nitrate could be interpreted as the consequence of massive disruption of endothelial integrity, with consequent passive nitric oxide release in vivo, or an adaptive mechanism of the endothelium to compensate for diffuse microvascular occlusion. The 2 mechanisms may both be involved, but the normal titers of nitric oxide end products in de novo TTP suggest that the former mechanism is more important, at least in cancer chemotherapy–related TTP.
Thrombotic thrombocytopenic purpura (TTP), a rare hematologic syndrome, is estimated to occur at the rate of 1 case per 1 million people.
1
TTP is characterized by occlusive microangiopathy preferentially localized to terminal arterioles and capillaries (but usually not venules) throughout the body. This typical histopathologic finding causes the classic clinical manifestations of TTP: microangiopathic hemolytic anemia, thrombocytopenia, fluctuating central nervous system abnormalities, fever, and renal impairment of varying degrees,2
although this classic pentad is observed in only 40% of TTP patients.The key pathogenic feature of TTP is the formation of platelet aggregates; however, the cause of such aggregates has been elusive and controversial for years. A substantial amount of experimental evidence points to an abnormal interaction between damaged vascular endothelium and platelets.
3
4
Indeed, the hypothesis of endothelial damage is supported by several experimental findings—abnormal production and metabolism of von Willebrand factor (vWF) multimers along with unusually large vWF multimers, which are capable of increasing platelet adhesiveness in vitro, shed into the circulation by endothelial cells4
5
; reduced vascular prostacyclin production6
; impaired fibrinolytic activity7
; increased vascular endothelial cell markers in blood8
; and proapoptotic effects of TTP plasma recently documented on microvascular endothelial cells in vitro.9
However, the initial cause of the primitive microvascular endothelial cell injury is still unclear, especially in the vast majority of TTP cases, the idiopathic ones.The occurrence of TTP in cancer patients has been well established,
10
, 11
, 12
although the relative importance of the underlying neoplasm and of some of the chemotherapeutic agents used, especially mitomycin,12
has not yet been fully clarified. The endothelial-damaging properties of several chemotherapeutic agents are well known13
, 14
, 15
, 16
, 17
, 18
as is the proaggregating activity of several tumors, in both in vitro and in vivo models.19
, 20
, 21
In this article, we describe 3 cases of TTP that developed in cancer patients receiving different chemotherapeutic regimens. In addition, we report further evidence of endothelial cell damage in these patients; extremely high levels of nitric oxide were indirectly detected in all of them, a finding different from what has been observed in patients with idiopathic TTP.
22
Report Of Cases
Three cancer patients in whom an episode of acute TTP developed during their neoplastic disease have been observed at our institutions since 1990. In these patients, blood samples were obtained before any treatment was initiated for their hematologic disease, and separated plasma was stored for subsequent biological analysis.
Case 1
A 50-year-old man with a silent familial and remote pathologic history of cancer began experiencing progressive dysphagia and pain during swallowing in the spring of 1990. Gastroscopy demonstrated a stenosing gastric cancer arising from the gastroesophageal junction. A biopsy performed during endoscopy revealed a gastric adenocarcinoma. A subsequent abdominal computed tomographic (CT) scan showed enlarged juxtagastric lymph nodes and diffuse liver metastases, findings that precluded surgical resection. Therefore, systemic chemotherapy was initiated according to the FAM protocol
23
(fluorouracil, 600 mg/m2 intravenously on days 1, 8, 29, and 36; Adriamycin [doxorubicin hydrochloride], 30 mg/m2 intravenously on days 1 and 29; and mitomycin, 10 mg/m2 intravenously on day 1); the cycles were repeated every 8 weeks. The patient received 3 complete cycles of full-dose chemotherapy as an outpatient, experiencing only grade 2 diarrhea and leukopenia during the second cycle. Two days after completing the third cycle, he was hospitalized because of a fever (38.5°C) and a diffuse petechial rash on his trunk and limbs. His neurologic condition worsened quickly, and eventually he was in an overt coma.On admission of the patient, blood test results showed severe thrombocytopenia (platelet count, 9 × 10
9
/L) associated with anemia (hemoglobin, 8.9 g/dL) but no granulocytopenia or neutropenia. Blood and urine cultures were negative. Coagulation tests demonstrated no fibrinogen degradation products, and prothrombin time, partial thromboplastin time, and fibrinogen values were within the reference range. Renal function was not impaired (creatinine level, 1.1 mg/dL), whereas the lactate dehydrogenase (LDH) level was substantially increased at 5791 U/L. Brain CT scan ruled out central nervous system metastases. TTP was diagnosed on the basis of several schistocytes found on a peripheral blood smear and the absence of both antierythrocyte and antiplatelet antibodies, which was subsequently confirmed by further immunohematochemical tests.Two plasma exchanges were promptly performed within the first 2 days of diagnosis; no cryosupernatant was used, and 1 volume of the patient's plasma was exchanged with the same amount of fresh frozen plasma without albumin. High-dose methylprednisolone was administered intravenously. Nevertheless, the patient's condition worsened, and he died 3 days after treatment had been instituted.
Case 2
A 60-year-old woman with a history of smoking (about 10 cigarettes per day since age 30 years) had a silent familial and remote history of cancer. After an episode of acute bronchitis in the fall of 1990, she began experiencing hemoptysis. A few weeks later, the symptoms persisted, and she underwent chest radiography, which demonstrated a central nodule in the left lung. CT-guided fine-needle biopsy revealed a pulmonary squamous cell carcinoma that showed signs of lymphatic spread and diffuse small metastases to the contralateral lung, findings that precluded surgery. Thus, the patient underwent systemic chemotherapy according to the PACCO regimen
24
(cisplatin, 50 mg/m2; Adriamycin, 50 mg/m2; cyclophosphamide, 300 mg/m2; and vincristine, 1.2 mg/m2; all were given intravenously on day 1 of each cycle in combination with lomustine [CCNU], 50 mg/m2, orally also on day 1, every 5 weeks). After the 3 cycles of chemotherapy, the patient experienced severe toxic effects, and the disease progressed. Therefore, she underwent second-line chemotherapy with use of a single agent, mitomycin (15 mg/m2 intravenously on day 1 every 4 weeks).25
One day after the administration of the second dose of mitomycin, the patient complained of fever and malaise; she had mild leukopenia (leukocyte count, 2.8 × 109
/L), no anemia (hemoglobin level, 12.6 g/dL), and a moderately low platelet count (117 × 109
/L). Two days later, she had a petechial rash on her trunk and legs and a persistent headache.On admission of the patient, hematochemical tests showed severe thrombocytopenia (platelet count, 14 × 10
9
/L) and anemia (hemoglobin level, 8.1 g/dL) with signs of intravascular microangiopathic hemolysis (LDH level of 1211.U/L, low levels of haptoglobin, and numerous schistocytes on a peripheral blood smear). Coagulation test results were normal, creatinine level was within the reference range (0.8 mg/dL), and no antierythrocyte antibodies were detected. Methylprednisolone therapy was initiated. Six plasma exchanges were performed in 10 days; no cryosupematant was used, and 1 volume of the patient's plasma was exchanged with fresh frozen plasma without albumin. On day 5, aspirin and dipyridamole were added, but her condition did not improve. She died of massive gastric hemorrhage.Case 3
A 62-year-old woman had been a heavy smoker (about 30 cigarettes per day) since the age of 17 years. She had a history of chronic bronchitis with typical seasonal phases characterized by episodes of productive cough, usually treated with antibiotics and mucolytics. In the spring of 1994, she began experiencing persistent irritative cough resistant to the antibiotics and mucolytics; the onset of fever, mainly in the evening (her temperature never exceeded 37.5°C), followed. She lost about 10 kg in 3 months. A few months later, she detected a soft mass in the skin of the right frontoparietal region of her skull; it was neither painful nor inflamed. It grew slowly but continually.
In the fall of 1994, the patient sought medical assessment because of persistent pulmonary symptoms. Chest radiography showed an expansive mass in the midlobe of her right lung, with ipsilateral hilar adenopathies. A staging CT scan showed several liver metastases and a skin metastasis, which was responsible for the growing skull mass, deeply infiltrating the skull and beginning to erode it. On the basis of needle biopsies of the skull metastatic lesion and the lung, a pulmonary adenocarcinoma was the likely diagnosis.
Palliation combination chemotherapy based on the CAP regimen
26
was initiated (400 mg/m2 of cyclophosphamide, 40 mg/m2 of Adriamycin, and 40 mg/m2 of cisplatin, all administered intravenously on day 1, every 4 weeks). The day after the completion of cycle 3, the patient had sudden development of a diffuse purpuric rash, followed by progressive sensory clouding—psychomotor impairment, hyperreactivity, and sleepiness. Emergency hemochrome analysis showed severe thrombocytopenia (platelet count, 19 × 109
/L), mild anemia (hemoglobin level, 9.3 g/dL), but no signs of granulocyte toxicity (leukocyte count, 3.6 × l09
/L).Thrombophilic diathesis or disseminated intravascular coagulation (DIC) was suspected. Hematochemical tests showed the following: prothrombin time, 12.6 seconds; partial thromboplastin time, 20.3 seconds; fibrinogen level, 238 mg/dL; and no fibrinogen degradation products. Several schistocytes but no other gross morphologic abnormalities of circulating blood cells were found on a peripheral blood smear. Further testing showed mild renal impairment (creatinine level, 1.3 mg/dL; blood urea nitrogen, 60 mg/dL), macrohematuria, 18.5% reticulocytes, LDH of 2657 U/L, and undetectable levels of haptoglobin. Results of a Coombs test were normal. TTP was diagnosed on the basis of the clinical pattern and the laboratory findings. The patient immediately underwent fresh frozen plasma infusion while we planned the necessary sessions of plasma exchange. A few hours later, however, the patient died of massive cerebral hemorrhage.
Biological Study
The high-performance liquid chromatographic (HPLC) technique that we used to detect stable nitric oxide end products, nitrite and nitrate, in the plasma has been described extensively.
22
Briefly, all blood samples were withdrawn early in the morning with use of a heparinized syringe; the patients had fasted overnight. Blood specimens were centrifuged at 2400g for 10 minutes at room temperature, and plasma samples were separated and frozen at −80°C until analysis.For HPLC analysis, a Supelcosil LC-18 column and a Supelguard LC-18 precolumn (Supelco Inc, Bellafonte, Pa) were used. HPLC factors were as follows: eluent, N octylamine 0.01 mol/L (adjusted to a pH of 6 with sulfuric acid); flow rate, 1.2 mL/min; detection UV, 205 nm; while retention times for nitrite and nitrate were 9.3 minutes and 15.1 minutes, respectively. Finally, the detection limits were 2 µmol/L for nitrite and 7 µmol/L for nitrate.
As expected, no nitrite was detected in the plasma of the 3 patients. As a matter of fact, the degradation of the nitric oxide molecule in the blood starts with binding to heme; nitric oxide met by oxyhemoglobin is rapidly converted to nitrate ion, and thus, methemoglobin is a side product. Consequently, although nitric oxide is degraded to nitrite and nitrate in roughly equal amounts in media containing no hemoglobin, nitrate represents the final metabolite of nitric oxide in the blood.
27
Nitrate titers in our 3 patients with chemotherapy-related TTP were 184.30 (case 1), 218.28 (case 2), and 185.00 µmol/L (case 3). These values are substantially higher than those observed in 29 patients with idiopathic TTP (average, 25.86 µmol/L) (Table 1) and in 29 sex- and age-matched healthy control subjects (average, 24.23 µmol/L).
22
These values suggest a severe endothelial dysfunction in our oncologic patients in whom TTP developed.Table 1Nitrate Titers in Controls and Patients With Idiopathic TTP and Chemotherapy-Induced TTP
| Nitrate titers (μmol/L) | ||||
|---|---|---|---|---|
| TTP | No. of patients | Mean ±SD | Median | Range |
| Idiopathic 22 | 29 | 25.86 ± 2.90 | 20.86 | 7.66-67.68 |
| Controls | 29 | 23.70 ± 10.02 | 24.23 | 7.00-53.69 |
| Chemotherapy induced | 3 | 195.86 ± 19.41 | 185.00 | 184.30-218.28 |
* TTP = thrombotic thrombocytopenic purpura.
Discussion
The initial association of TTP or of its closely related nephrologic counterpart hemolytic-uremic syndrome (HUS) with antineoplastic drugs was in 1971 when Liu et al
28
reported that mitomycin-induced renal toxicity was associated with, among other features, microangiopathic hemolytic anemia and thrombocytopenia. Since then, several investigators have described chemotherapy-induced TTP.10
, 11
, 12
Mitomycin is the antiblastic agent most commonly considered responsible for TTP development in cancer patients,28
, 29
, 30
but other agents have been implicated, including bleomycin,31
cisplatin,32
lomustine,33
pentostatin (deoxycoformycin),34
several combinations,35
, 36
, 37
and even the antiestrogen tamoxifen (although through a supposed interaction with other drugs, including mitomycin).38
An increasing number of TTP/HUS cases have been reported after bone marrow transplantation, during induction treatments (supramaximal chemotherapy, irradiation, or both), or during posttransplantation immunosuppressive treatment, especially with cyclosporine (also with tacrolimus).
39
The prognosis of cancer chemotherapy-related TTP/HUS is particularly severe, as demonstrated in our patients. Treatment has usually been the same as that for de novo TTP/HUS—ie, plasma exchange and corticosteroids with or without antiplatelet drugs—even though the effectiveness of plasma immunoadsorption was recently reported.
30
Because patients receiving chemotherapy are being treated for malignancies, the relative contribution of the drugs or of the underlying cancer to the onset of TTP/HUS remains difficult to assess. This is particularly true considering how often cancer patients have hypercoagulability states or a frank DIC,
19
, 20
, 21
which should be immediately distinguished from TTP/HUS in a cancer patient in whom thrombocytopenic hemorrhagic diathesis has developed.12
The major distinction between TTP/HUS and DIC is the consumption of coagulation factors in DIC but not in TTP/HUS; however, this difference is sometimes difficult to assess in cancer patients because the level of coagulation factors is usually low, especially in those with advanced cancer in whom fibrin deposition in tumor tissue occurs frequently.40
Furthermore, because. extensive vascular endothelial injury can result in either DIC or TTP/HUS, both conditions may occur in the same patient.As previously stated, a substantial amount of evidence supports the hypothesis that the clinical manifestation of both idiopathic and cancer chemotherapy-related TTP/HUS is due to an insult to the vascular endothelium; thus, little reason exists to believe that the pathophysiologic process underlying the clinical development of TTP/HUS differs from that underlying idiopathic and cancer chemotherapy-related TTP/HUS. However, this was recently challenged. Nagaya et al
29
found that the levels of the cytokines tumor necrosis factor α, interleukin 1β, and interleukin 6 are increased in patients with de novo TTP/HUS and DIC but not in those with mitomycin-induced TTP/HUS. In contrast, vWF antigen and low-molecular-weight vWF multimers are decreased in de novo TTP/HUS but increased in DIC and in the drug-induced syndrome,29
findings that can help to distinguish among these 3 syndromes.We report another feature of cancer chemotherapy-related TTP, which is absent in de novo TTP cases—the increased release of nitric oxide into the bloodstream, probably from the chemotherapy-injured vascular endothelium. Under physiological conditions, the shear stress–induced release of nitric oxide continuously causes the relaxation of vascular smooth muscle cells and contributes to the prevention of platelet adhesion and aggregation in normal blood vessels. Furthermore, substances with strong vasoconstrictor properties can induce nitric oxide release and thus counteract their action.
41
Even though an altered nitric oxide release might be postulated in TTP, a normal nitric oxide production was recently demonstrated in patients with de novo TTP.22
Thus, other pathophysiological mechanisms may be more important in these patients. Indeed, at the endothelial level, different substances contribute to normal endothelial homeostasis; for example, prostacyclin is one of these substances that has been implicated in TTP pathogenesis.42
In our 3 cancer patients in whom TTP developed after treatment with antiblastic drugs (whether or not they contained mitomycin), the observed increased release of nitric oxide could be interpreted as the consequence of massive disruption of endothelial integrity due to the drugs acting directly on the endothelium, to the underlying malignancies, or both, with consequent passive release of nitric oxide in vivo. Several antineoplastic agents can induce an increase of nitric oxide stable end products, although to a different extent43
(C.P., unpublished observation, 1997). However, an adaptive mechanism with which the endothelium tries to compensate for diffuse microvascular occlusion could also be postulated. Finally, activated leukocytes could be the source of the observed increased nitrate titers. These mechanisms may also be involved together, but the normal titers of nitric oxide end products in de novo TTP suggest that the massive disruption of endothelial integrity is the most important, at least in cancer chemotherapy-related TTP. Our series was small, but this condition is rare. Further studies are needed with larger series and titration of more samples during the course of disease.References
- Thromboiic thrombocytopenic purpura: a thirty year review.Semin Thromb Hemosl. 1980; 6: 350-355
- TtiTombotic thrombocytopenic purpura: report of 25 cases and review of the literature.Medicine (Baltimore). 1981; 60: 413-428
- The changing couisc of thrombolic thrombocytopenic purpura and modem therapy.Blood Rev. 1993; 7: 94-103
- Studies on the pathophysiology of thrombolic thrombo-cytopenic purpura.Semin Hemalol. 1997; 34: 83-89
- Abnormalities of von Willebrand factor multimers in drug-associated throtnbotic micro-angiopathies.Am J Hematol. 1993; 42: 268-277
- Prastacyclin deficiency in thrombolic microangiopalhy [letter].Lancet. 1981; 2: 1422-1423
- Clinicopathologic features of thrombolic thrombocytopenic purpura.Semin Hematol. 1987; 24: 71-81
- Increased levels of vascular endothelial cell markers in thrombolic thrombocytopenic purpura.Am J Hematol. 1993; 44: 101-105
- Plasma from patients with idiopathic and human immunodeficiency virus-associated thrombolic thrombocytopenic purpura induces apoptosis in micro vascular endothelial cells.Blood. 1996; 87: 3245-3254
- Microangiopathtc hemolytic anemia and cancer; a review.Medicine iBallimore). 1979; 58: 377-384
- Thrombolic microangiopalhy in the cancer patient including those induced by chemotherapeutic agents.Semin Hematol. 1987; 24: 161-177
- Cancer- and drug-associated throtnbotic thrombocytopenic purpura and hemolytic uremic syndrome.Semin Hematol. 1997; 34: 140-147
- Chemotherapy enhances endothelial cell reactivity to platelets.Clin Exp Metastasis. 1990; 8: 511-518
- Influence of mitomycin C on endothelial monolayer regeneration in vitro.J Cell Biochem. 1992; 50: 293-300
- Toxicity of mitomycin C toward cultured pulmonary artery endothelium.Toxicol Appl Pharmacol. 1995; 130: 87-94
- Effects of anticancer drugs on vascular wall: histological and pharmacological studies.Nippon Igaku Hoshasen Gakkai Zasshi. 1996; 56: 818-821
- The influence of S-fluorouracil and methotrexate on vascular endothelium: an experimental study using endothelial cells in the culture.Ann Oncol. 1996; 7: 731-737
- Amifostine (WR-2721, Ethyol(tm)) can protect human endothelial but not tumor cells from cisplatin-induced cytotoxicity: an in vitro study.Oncol Rep. 1997; 4: 729-732
- Proaggregating and procoagulant activities of human mesothelioma tumor cells at different stages of “in vitro” culture.Haemaiologica. 1991; 76: 392-397
- Effect of different platelet agonists on intracellular free Can concentrations in human tumor cells: possible role in tumor growth.Int J Cancer. 1995; 62: 291-296
- Platelets and cancer metastasis: more than an epiphenomenon.Semin Thromb Hemost. 1992; 18: 392-415
- Nitrite and nitrate plasma levels, as markers for nitric oxide synthesis, in thrombotic thrombocytopenic purpura (TTP): a case-control study.Hematology. 1996; 1: 239-246
- 5-Fluorouracil, Adriamycin, and mitomycin-C (FAM) combination chemotherapy in the treatment of advanced gastric cancer.Cancer. 1979; 44: 42-47
- cis-Dichloro-diammineplatinum(II), Adriamycin, cyclophosphamide, CCNU, and vincristine in non-small cell lung carcinoma: a preliminary report.Cancer Treat Rep. 1979; 63: 29-33
- The case for mitomycin in non-small cell lung cancer.Oncology. 1993; 50: 35-50
- Differing response rates and survival between squamous and non-squamous non-small cell lung cancer: comparison of CAP versus MAP.Am J Clin Oncol. 1986; 9: 249-254
- Endothelial nitric oxide and cardiovascular disease.J Intern Med. 1994; 235: 317-327
- Renal toxicity in man treated with mitomycin C.Cancer. 1971; 28: 1314-1320
- Hemosiatic abnormalities and increased vascular endothelial cell markers in patients with red cell fragmentation syndrome induced by mitomycin C.Am J Hematol. 1995; 50: 237-243
- Treatment of cancer chemotherapy-associated thrombotic thrombocytopenic purpura/hemolytic uremic syndrome by protein A immunoadsorption of plasma.Cancer. 1993; 71: 1882-1892
- Thrombotic micro-angiopathy and renal failure associated with antineoplastic chemotherapy.Ann Intern Med. 1984; 101: 41-44
- Cisplatin-associated hemolytic uremic syndrome.Cancer. 1994; 74: 3059-3062
- Chronic glomcrular microangiopalhy and metastatic carcinoma.HwnPathol. 1979; 10: 433-438
- Hemolytic-uremic syndrome due to deoxycoformycin: a report of the second case.Intern Med. 1995; 34: 593-596
- Elevated plasma von Willebrand factor levels and arterial occlusive complications associated with cisplatin-based chemotherapy.Oncology. 1985; 42: 296-300
- Renal vascular lesions after chemotherapy with vinblastine, bleomycin, and cisplatin.Am J Med. 1982; 73: 429-433
- Thrombotic thrombocytopenic purpura subsequent to acute myelogenous leukemia chemotherapy.Am J Hematol. 1986; 21: 299-304
- A toxic interaction between mitomycin C and tamoxifen causing the haemolyttc uraemic syndrome.Ear J Cancer. 1993; 29A: 1854-1857
- Transplantation-associated ihrombotic thrombocytopenic purpura and hemolytic uremic syndrome.Semin Hematol. 1997; 34: 126-133
- Localization of intracranial tumors: comparative study with mI-labeled antibody lo human fibrinogen and neohydrin-w,Hg.Cancer. 1967; 20: 751-755
- Endothelium-derived relaxing and contracting factors.WienKlin Wochenschr. 1991; 103: 405-411
- Role of prastacyclin in the pathogenesis and therapy of thrombotic thrombocytopenic purpura.in: Kaplan BS Trompeter RS Moake JL Hemolytic Uremic Syndrome and Thrombotic Thrombocytopenic Purpura. Marcel Dekker Inc, New York, NY1992: 483-489
- Nitrate plasma level, as a marker of nitric oxide production after subcutaneous interleukin 2 immunotherapy [letter].JNall Cancer Inst. 1997; 89: 1545
Article Info
Identification
Copyright
© 1999 Mayo Foundation for Medical Education and Research. Published by Elsevier Inc. All rights reserved.
