C-Reactive Protein Monitoring Can Predict Neutropenic Fever during Autologous Hematopoietic Stem Cell Transplantation for Mutliple Myeloma

Background

Post-auto transplantation in multiple myeloma, neutropenic infections typically occur earlier than engraftment.It may be difficult to distinguish between infection and other causes of neutropenic fever, such as engraftment syndrome.Early indication of a neutropenic fever helps to identify those patients who are more likely to benefit from a combination of antibiotics, while reducing unnecessary toxicity in patients with a low probability of a neutropenic fever. C-reactive protein (CRP) is an acute phase protein and an important marker of inflammation that is elevated with serious infections, but also with high levels of certain cytokines as typically seen post-transplantation. We retrospectively analyzed the role of serial serum CRP levels in predicting neutropenic fever.

Materials and Methods

Patients included received auto transplants (first and/or second) for multiple myeloma between January 2014 and March 2017. IRB approval was obtained. One hundred and seventy consecutive transplants received by 121 patients were included. Serum CRP levels were measured regularly during transplant and those obtained between Day 0 and Day 14 post-transplant were analyzed. Patients were started on prophylactic Meropenem on day +5 post-transplant. To assess the prognostic effect of C-reactive protein on onset of neutropenic fever within 14 days post-transplant, a Cox regression model was applied. Time was calculated from transplant to onset of neutropenic fever, defined as neutropenia (ANC<500/µl or when ANC is expected to fall below 500 over the next 48 hrs.) and fever (one temperature reading > 101.0⁰ F or at least two temperature readings > 100.4⁰ F). CRP was analyzed as a time-dependent covariate. A robust sandwich estimator of the standard error was utilized to account for a clustering effect of multiple transplants within a patient. Estimated effects were reported as hazard ratios (HR) along with 95% confidence intervals (95% CI). Additionally, area under the curve and maximum CRP value (either before or after neutropenic fever) were compared between groups using mixed effects regression models.

Results

Of the 170 transplants, 49 developed neutropenic fever and 121 did not have neutropenic fever. On multivariable analysis, high levels (4.0-42.3 mg/dl) of CRP were associated with a 5.45 times increased risk of neutropenic fever (p-value = 0.02) compared to undetectable levels (<0.5 mg/dl). Patients had a nearly 3 times increased risk of neutropenic fever after the second transplant (p-value<.01). Area under the curve for consecutive CRP values was larger for those with neutropenic fever (p-value<.01). Those with neutropenic fever had higher max values of CRP (p-value<.01) which tended to occur after the onset of neutropenic fever.

Conclusion

CRP monitoring may provide important information about the risk for neutropenic fever during autologous hematopoietic stem cell transplantation for Multiple Myeloma. Serial CRPs may also help identifying low-risk patients, who can continue outpatient treatment. Therefore, in our opinion, CRP levels should be part of the daily routine laboratory workup post-autotransplants.