December 10, 2009
5 min read

Recognizing gemcitabine-induced thrombotic thrombocytopenic purpura

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

With the continual growth of the number of cytotoxic and targeted chemotherapeutics, as well as novel combinations of different agents, it is more crucial than ever that treating oncologists are aware of relatively rare, but serious adverse effects. This need is exemplified by a case that we encountered at our institution earlier this year.

A 46-year-old white woman presented to the hospital with a primary complaint of persistent nausea, vomiting and diarrhea, four days after beginning her fourth cycle of a phase-2 clinical trial of gemcitabine and imatinib (Gleevec, Novartis).

Amit Mehta, MD
Amit Mehta

Eight years prior to the current admission, the patient was diagnosed with left-sided invasive breast carcinoma, with both ductal and lobular features. At that time, she was found to have stage IIa breast carcinoma, which was positive for ER, PR and HER2/neu overexpression. Her initial management consisted of a left modified radical mastectomy, and she subsequently received chemotherapy with doxorubicin and cyclophosphamide, followed by paclitaxel. After this, she began tamoxifen and was eventually switched to letrozole. Genetic testing revealed that she had a BRCA1 germline mutation. Due to this, the patient also elected for a prophylactic contralateral mastectomy, as well as a bilateral oopherectomy.

The patient subsequently developed metastatic disease to the bone six years after her initial diagnosis. Multiple endocrine therapy manipulations were attempted, as well as a course of capecitabine (Xeloda, Roche), but the bone disease ultimately progressed. At this point in her course, she had enrolled into the phase-2 study of a combination of gemcitabine and imatinib.

The first three cycles of this treatment were well-tolerated, other than some mild nausea. Her only medications were bupropion, omeprazole and ondansetron as needed. Along with the presenting gastrointestinal symptoms, she also had a fever of 101oF with chills and cramping abdominal pain. Upon admission, her physical examination was remarkable for mild tachycardia and subtle scleral icterus. Initial laboratory data revealed a mild hyperbilirubinemia and transaminase elevations (see table 1).

Patient History

Based on her initial assessment, she was managed conservatively with hydration. Blood cultures, urine culture and Clostridium difficile toxin assay were negative. At first, the unexpected elevation in bilirubin was attributed to possible gemcitabine toxicity, but the cause was not entirely clear. However, throughout the next few days, her laboratory parameters changed dramatically (see table 2).

The substantial decrease in both the hemoglobin and platelet count, combined with the marked rise in LDH, prompted an urgent evaluation of the peripheral blood film. This revealed approximately 10 to 12 schistocytes per high-power field.

A hemolytic panel showed that the haptoglobin was depressed at less than 7 mg/dL, and Coomb’s testing was negative. Coagulation parameters on hospital day six revealed normal prothrombin and partial thromboplastin times. The fibrinogen was 591 mg/dL, with fibrin degradation products intermediate-positive at five to 20. Stool cultures were negative for E. Coli 0157:H7 and other organisms.

In light of these results, along with the serum creatinine tripling in value and the indirect bilirubinemia of 6.6 mg/dL, the clinical diagnosis of thrombotic thrombocytopenic purpura was made. Arrangements were expeditiously made for the placement of a Shiley catheter, and daily plasmapheresis was initiated.

Over the next few days, the platelet count and hemoglobin stabilized, and by hospital day 12, the hemoglobin rose to 8.7 g/dL, and the platelet count to 73,000 /mcL. Still, with the LDH ranging from 1,065 IU/L to 1,238 IU/L, and a total bilirubin of 9.1 mg/dL, the microangiopathic hemolytic anemia was quite active. Plasmapheresis was continued, along with supportive packed red blood cell transfusions as needed.

During this time, however, the creatinine rose to 4.3 mg/dL, and the patient became anuric. As a consequence, daily hemodialysis was begun. In addition, the patient developed disseminated intravascular coagulation, presumably secondary to TTP.

The fibrinogen trended down to 342 mg/dL and fibrin degradation products to greater than 20. The prothrombin time became slightly prolonged to 15 seconds, but fortunately there was no clinical bleeding.

The treatment at this point consisted of continuing plasma exchange, as well as hemodialysis, on a daily basis. Hemolytic parameters, including the LDH, bilirubin and CBC, were monitored regularly, as was the peripheral smear for schistocytes. Blood products were given supportively as needed, but platelet transfusions were avoided due to the presence of TTP.

Seven days after beginning plasmapheresis, the platelet count finally increased to more than 100,000 /mcL. However, the reticulocyte count remained at 12% to 15%, with an LDH value of approximately 500 IU/L. The patients occasionally required blood transfusions as well and remained anuric. Hence, daily plasmapheresis and daily hemodialysis were continued.

After another week, the CBC and hemolytic parameters began to normalize. Transfusions were no longer needed, and the LDH dropped to about 200 IU/L. At this point, the frequency of exchanges was decreased to every other day, and the parameters still remained stable. Soon afterwards plasma exchange was discontinued.

Much to the joy of the patient and the medical team, the patient also began to recover adequate kidney function. In fact, she did not require any further dialysis postdischarge. The patient continues to do well about five months later.

Laboratory Data During the Initial Week of Hospitalization

Case Discussion

Given the clinical picture, as well as the new chemotherapeutic exposure, the TTP was most likely secondary to gemcitabine. The diagnosis of TTP was made on clinical and laboratory grounds, and treatment was begun promptly. Interestingly enough, however, the ADAMTS-13 enzyme assay showed a level of 39%.

Typically, TTP is associated with severe deficiencies of ADAMTS-13, with values less than 5%. However, as the literature for gemcitabine has grown with its increasing usage, it has been noted that only about 13% of patients with gemcitabine- associated TTP feature such extreme deficiencies. This implies that clinical recognition of this entity is even more paramount.

TTP associated with gemcitabine has a reported incidence between 0.015% and 1.4%. In addition, the risk of this complication appears to correlate with a cumulative dose of gemcitabine of approximately 20,000 mg/m2, but multiple case reports in the literature recount exceptions.

In addition, as more novel combination therapies are tested, such as with the current patient, unknown interactions may be occurring that could potentiate such adverse effects.

The mechanism for microangiopathic hemolytic anemia with this chemotherapeutic agent has not been fully elucidated. The root pathway for this process is thought to involve endothelial damage, due to the finding of similar pathologic changes with other agents associated with TTP. However, in concert with the discrepancy in ADAMTS-13 levels in comparison to idiopathic TTP, it does stand to reason that the mechanism for microangiopathic injury is not the same between gemcitabine-associated and idiopathic etiologies.

Regardless of the exact mechanism, the optimal treatment approach appears to be the same. As was employed in the current case, plasma exchange along with drug discontinuation are the mainstays of therapy. Several published reports show that patients can eventually recover normal renal function, although they may need dialysis support until the thrombotic microangiopathy resolves.

In conclusion, as gemcitabine is utilized in the treatment algorithm of a greater diversity of malignancies, and as a component of novel combinations, it is crucial that oncologists remain cognizant of the association with TTP.

The diagnosis is made on clinical grounds, with proper support of the peripheral blood film and relevant laboratory parameters. Urgent therapy is critical and can make a substantive impact in the morbidity and mortality of a patient.

Amit Mehta, MD, is a third-year Fellow in the Division of Hematology and Oncology at University of Medicine and Dentistry, New Jersey, and is a member of the HemOnc Today Editorial Board.

For more information:

  • Flombaum CD. Am J Kidney Dis. 1999;3:555-562.
  • Moake JL. N Engl J Med. 2002;347:589-600.
  • Walter RB. Am J Kidney Dis. 2002;40:E16.
  • Zupancic M. Lancet Oncology. 2007;8:634-641.