Hwa, Lisa

ABSTRACT Ciltacabtagene autoleucel (cilta‐cel) has demonstrated excellent efficacy and long‐term disease control in patients with relapsed/refractory multiple myeloma (RRMM). However, a proportion of patients experience early treatment failures. We investigated clinical factors associated with early progression or death (within 12 months) in patients with RRMM receiving standard‐of‐care cilta‐cel across the three Mayo clinic centers. Patients with a follow‐up of at least 12 months or progression or death within 12 months from infusion were included. Of the patients with early treatment failure ( n = 52), 69% had progressive disease and 31% had a non‐relapse mortality (NRM) event. In patients without early treatment failure ( n = 164), 13% of events were NRM. Among pretreatment factors, prior BCMA‐directed therapy, presence of extramedullary disease and a CAR‐HEMATOTOX score of ≥ 2 were independent predictors of early progression or death. The utilization of cilta‐cel in earlier lines of treatment (1–3 vs. 4 or more) demonstrated comparable PFS (12‐month PFS 73% vs. 77%, p = 0.94). Measurable residual disease positivity in the bone marrow at 3 months (11/197 patients) identified a small but high‐risk group with an increased risk of early treatment failure (OR 5.68; p = 0.012). Patients with less than a complete response on a FDG PET/CT at 3 months also had an increased risk of early treatment failure (OR 11.8; p < 0.0001). Our findings may help identify patients at high‐risk for early adverse outcomes despite receiving highly effective therapy, and support consideration of novel therapeutic strategies within a clinical trial setting.

Abstract Immune effector cell–associated enterocolitis (IEC-EC) has been observed after ciltacabtagene autoleucel (cilta-cel [Carvykti]) chimeric antigen receptor T-cell (CAR-T) therapy. It is associated with dismal outcomes and is poorly characterized. Here, we examined its clinical features, risk factors, and proposed management strategies. Among the 229 consecutive patients who received cilta-cel at Mayo Clinic, 9 (3.9%) presented with a nonresolving, nonbloody, grade 3 diarrhea, often requiring prolonged total parenteral nutrition. Gastrointestinal coinfections were seen in most patients. Median time from CAR-T infusion to symptom onset was 85 days (range, 35-166), and there appeared to be a latency from symptom onset to endoscopic evaluation. Histopathology findings resembling a graft-versus-host disease pattern of mucosal injury were most seen on duodenal biopsies, whereas 2 cases had patterns akin to a T-cell lymphoproliferative disorder (1 CD4+ and 1 CD8+). Independent associated factors for IEC-EC included high-risk disease as defined by the International Myeloma Society and International Myeloma Working Group, prolonged cytokine release syndrome, and antecedent delayed neurotoxicity. Response to first-line therapy comprising IV or oral corticosteroids, bile acid sequestrants, IV immunoglobulin, and antimicrobial-directed therapy was poor, with no durable responses. Biologic therapy appeared to induce durable responses in 3 of the 5 patients (2/2 vedolizumab; 1/3 infliximab); none of the responders had gastrointestinal coinfection. Early gastroenterology input for endoscopic evaluation with the inclusion of duodenal biopsy is a key component for diagnosis. Early institution of biologic therapy after failure of a short course of corticosteroids should be considered.

Abstract Gastrointestinal complications of multiple myeloma (MM) treatment are common and include nausea, constipation, and diarrhea. However, acute gastrointestinal events like perforations are rare. We aimed to describe the characteristics and outcomes of patients with MM that had colonic perforations during their treatment. This is a retrospective study that included patients from all three Mayo Clinic sites who had MM and developed a colonic perforation. All patients were diagnosed with colonic perforations based on CT scans and were surgically treated. Patients diagnosed with AL amyloidosis, a perforated colon complicating neutropenic colitis during ASCT and those with perforation due to colonic cancer were excluded. A high dose of dexamethasone was defined as ≥40 mg dexamethasone once a week. Thirty patients met inclusion criteria. All patients received steroids at doses ≥10 mg once weekly prior to the perforation, while four (11%) were on high‐dose dexamethasone without chemotherapy. Fourteen patients were given high doses of dexamethasone. Twenty‐five patients required ostomies with all surviving surgery. Twenty‐four perforations (80%) were associated with diverticulitis. Treatment with steroids was resumed in 23 patients with no further gastrointestinal complications. The median OS was 20 months following perforation (IQR 8–59). Within the same timeframe 5854 patients were treated at Mayo Clinic for MM, making the risk of bowel perforation 0.5%. Intestinal perforations in MM are rare and, in our series, always occurred with dexamethasone ≥10 mg per week. Urgent surgery is lifesaving and resumption of anti‐myeloma treatment appears to be safe.

AbstractDetection of myeloma progression (PD) relies on serial 24‐h urinary M protein measurements in patients without measurable serum M spike. We examined whether serial difference free light chain (dFLC) levels could be used as a surrogate for serial 24‐h urine M protein measurements in monitoring for PD in patients with baseline measurable urine M protein. We studied 122 patients who had serial measurement of urine M protein and serum FLC and had demonstrated PD. The median increase in dFLC with progression as defined by urine M spike was 110% (IQR: 55‐312) and median absolute increase was 74 mg/dL; while 89% of patients had dFLC increase ≥ 25%, 94% had absolute increase in dFLC > 10 mg/dL, and 98% met at least 1 of these 2 criteria at PD. In patients with baseline measurable serum FLC (n = 118), 89% had increase in dFLC ≥ 25%, 97% had dFLC increase of > 10 mg/dL, and 98% had 1 of the 2. We conclude that serial dFLC assessments can be used in place of serial 24‐h urine protein assessments during myeloma surveillance to monitor for PD. Once patients have an absolute increase in dFLC of >10 mg/dL from the nadir, a 24‐h urine collection can then be assessed to document PD as per the International Myeloma Working Group criteria.

Over the past decade, use of novel agents, including immunomodulatory drugs (IMiDs) and proteasome inhibitors (PIs) has resulted in high response rates and improvement in overall survival (OS) for patients with multiple myeloma (MM); however, the prognostic significance of refractoriness to these agents when used as initial therapy has not been extensively studied. We reviewed the outcomes of 816 consecutive patients treated for MM at our institution since 2006 to evaluate the survival difference between those achieving at least a partial response (PR) to induction therapy and those who were primary refractory. The median OS from start of therapy was significantly shorter for the primary refractory group at 3.6 vs. 7.6 years for the responding patients ( P < 0.001). The difference in median OS persisted when only patients receiving a novel agent as part of induction therapy were considered (3.6 vs. 7.9 years, P < 0.001) and in a 4‐month landmark analysis (4.2 vs. 7.6 years, P < 0.001). The median OS for patients achieving a complete response (CR), very good partial response (VGPR), PR, or less than PR was not reached (NR), 6.1, 6.4, and 4.2 years from the 4‐month landmark, respectively ( P < 0.001). The comparatively poor outcomes of patients refractory to induction therapy in the current era of novel agents suggests that this high‐risk subpopulation must be further studied for predictors of resistance and, when identified, should be targeted for clinical trials. Am. J. Hematol. 90:981–985, 2015. © 2015 Wiley Periodicals, Inc.

The use of soluble cardiac biomarkers such as N‐terminal pro‐brain natriuretic peptide (NT‐proBNP) and troponin has revolutionized prognostication for patients with AL amyloidosis. Soluble ST2 (sST2) and galectin‐3 have also been reported to have prognostic value in other cardiac patient populations. We identified 502 patients with AL amyloidosis, who provided a research sample and consent to review their medical records between 1/1/2006‐12/31/2010 within 90 days of their diagnosis. Samples were assayed for sST2 and galectin‐3. Within this AL amyloidosis population, overall survival (OS) was 25.5 months (95% CI 18, 35.7 months). Receiver operating curve analyses were done to detect the best cut‐points for sST2 and galectin‐3 to predict both 1‐ and 5‐year OS. The respective cut points for sST2 were 30 and 29.7 ng/mL, while the median sST2 for the entire population was 31 ng/mL (IQR 19.8, 53.6). The respective cut points for galectin‐3 were 11 and 10.4 ng/mL while the median for the entire population was 16.6 ng/mL (IQR 11.5, 24.0). Although on univariate analysis, both sST2 and galectin‐3 were prognostic, upon multivariate analysis, only sST2 was independent of troponin, NT‐proBNP, serum immunoglobulin free light chain, and blood pressure. Not only did sST2 add to previously reported prognostication systems, but a novel prognostication 5‐point system including sST2 was possible. The addition of sST2 – but not galectin‐3 – to existing prognostication systems for patients with AL amyloidosis strengthens the ability to predict for death. Am. J. Hematol. 90:524–528, 2015. © 2015 Wiley Periodicals, Inc.