Author: admin

Psychotherapy leads to measurable changes in brain structure. Researchers at Martin Luther University Halle-Wittenberg (MLU) and the University of Münster have demonstrated this for the first time in a study in “Translational Psychiatry” by using cognitive behavioral therapy. The team analyzed the brains of 30 patients suffering from acute depression. After therapy, most of them showed changes in areas responsible for processing emotions. The observed effects are similar to those already known from studies on medication.

Around 280 million people suffer from severe depression worldwide. This depression leads to changes in the brain mass of the anterior hippocampus and amygdala. Both areas are part of the limbic system and are primarily responsible for processing and controlling emotions. In psychotherapy, cognitive behavioral therapy (CBT) is an established method for treating depression.

CBT leads to positive changes in thought patterns, emotions and behavior. We assume that this process is also linked to functional and structural changes in the brain. The effect has already been demonstrated with therapy involving medication or electrostimulation, but has not yet been proven for psychotherapy in general.”

Professor Ronny Redlich, Head, Department of Biological and Clinical Psychology at MLU

Now researchers at MLU and the University of Münster have succeeded in demonstrating this in an extensive study involving 30 people suffering from acute depression. Structural magnetic resonance imaging (MRI) was used to examine the participants’ brains before and after 20 sessions of therapy. “MRI scans provide information about the size, shape and location of tissue,” explains psychologist Esther Zwiky from MLU. In addition to the MRI scans, clinical interviews were conducted to analyse the symptoms of the disease, such as difficulty in identifying and describing feelings. In addition, 30 healthy control subjects, who did not undergo therapy, participated in the study for comparison purposes.

The study provided clear results: 19 of the 30 patients were found to have hardly any acute depressive symptoms after therapy. The researchers were also able, for the first time, to document specific anatomical changes. “We observed a significant increase in the volume of grey matter in the left amygdala and the right anterior hippocampus,” says Esther Zwiky. The researchers found a clear connection to the symptoms: individuals with a greater increase in grey matter in the amygdala also showed a stronger reduction in their emotional dysregulation.

“Cognitive behavioral therapy was already known to work. Now, for the first time, we have a reliable biomarker for the effect of psychotherapy on brain structure. Put simply, psychotherapy changes the brain,” explains Ronny Redlich. However, Redlich stresses that there is no fundamentally better or worse treatment – medication works better for some people, while electrostimulation works very well for others; for others, CBT can be most helpful. “It is therefore all the more encouraging that we were able to show in our study that psychotherapy is an equally effective alternative from a medical and scientific standpoint,” says Redlich.

The study was supported German Research Foundation (DFG), the Federal Ministry of Research, Technology and Space (BMFTR) and the state of Saxony-Anhalt.

Nearly 80% of US neurologists prescribing drugs for multiple sclerosis (MS) received at least one pharma industry payment, with higher volume prescribers more likely to be beneficiaries, finds a 5 year analysis of Medicare database payments, published in the open access journal BMJ Open.

And those in receipt of these payments were more likely to prescribe that company’s drugs, especially if the sums involved were larger, sustained, and recent, the findings indicate.

Because of the lifelong nature of MS, effective therapies are usually continued indefinitely unless a patient’s clinical response changes, explain the researchers. And MS drug prescriptions are Medicare’s largest neurological drug expense despite making up a relatively small portion of total claims, they add.

While previously published research indicates that industry payments are associated with increased prescribing of marketed products, none of these studies focused on a market as competitive as the MS drugs market, say the researchers. 

They therefore set out to characterise industry payments to neurologists prescribing MS drugs and find out if the receipt of such payments might be associated with the likelihood of the preferential prescribing of that company’s drugs.

They used publicly available data on payments made by pharma companies to doctors from the Centers for Medicare & Medicaid (CMS) Open Payments platform from 2015 to 2019.

Payments are classified as: research payments; ownership and investment interests; and general payments. The researchers focused on general payments to neurologists, linking these to Medicare Part D data, which covers prescription drugs, using National Provider Identification numbers and drug names. 

Their analysis included 7401 neurologists who had prescribed disease modifying therapies (DMTs) for at least 1 year, issuing a minimum of 11 prescriptions, and 20 DMTs manufactured by 10 companies.

In all, 5809 (78.5%) neurologists received 626,290 distinct industry payments from at least one drug company, totalling US$163.6 million between 2015 and 2019; 4999 (67.5%) neurologists received payments from two or more companies.

The average individual amount received was US$779, but 10% of recipients amassed US$155.7 million between them-95% of the total sums received–which suggests that drug companies may selectively target high-volume prescribers, say the researchers.

Higher prescription volumes were associated with a greater likelihood of receiving any payment type, particularly for consulting services, non-consulting services, such as speaking at an event, and travel/accommodation; the highest number of discrete payments was made for food and drink. 

The amount received was positively associated with prescription volume. Compared with those who received no payments from a company, those who did, were 13% more likely to prescribe that company’s drugs.

The strongest association between industry payment and prescribing tendencies was observed for non-consulting services. These neurologists were 53% more likely to prescribe that company’s drugs. 

Larger payments were also associated with a greater likelihood of preferential prescribing, rising in tandem with the size of the payment received: US$50 was associated with a 10% greater likelihood of prescribing that company’s drugs; US$500 with a 26% greater likelihood; US$1000 with a 29% greater likelihood; and US$5000 with a 50% greater likelihood.

Longer duration of payments was another seemingly influential factor, ranging from a 12% greater likelihood of prescribing that company’s drugs for one year of payments to 78% greater likelihood for 5 consecutive years. 

The recency of payments also seemed to be influential. A payment made 4 years earlier was associated with a 3% greater likelihood of prescribing that company’s drugs, but a 34% greater likelihood when made in the same year.

This is an observational study, and as such, no firm conclusions can be drawn about cause and effect. And the researchers acknowledge that their study was limited to the prescribing of Part D drugs, and couldn’t establish the appropriateness of prescribing, nor for which patients more expensive brand-name drugs were most suitable. 

A doctor’s decision to prescribe is informed by many different factors, including national guidelines and/or institutional protocols, insurance cover, and patient preferences. These drivers are difficult to assess using publicly available data, but should be considered when interpreting the findings, emphasise the researchers.

Nevertheless their “findings raise concerns about excess pharmaceutical promotion efforts and their implications for physician prescribing for patients,” they suggest.

“Promotional efforts to influence prescribing are especially concerning given the drugs’ substantial costs, particularly if more expensive brand-name drugs are being prescribed instead of appropriate, effective, generically available alternatives,” they point out.

“The Physician Payments Sunshine Act, which led to the creation of the Open Payments Database, was an important step forward in making transparent the financial conflicts of interest among physicians receiving industry payments.

“However, it remains unclear whether increased transparency has mitigated these conflicts of interest and their impact on prescribing behaviour, or simply given the public greater insight into the large scale of industry payments made to prescribers,” they conclude.

In a large case-control study with thorough control of major confounders, we did not observe associations between lung cancer risk and air pollution values estimated as annual average for the year 1990, i.e., 12–15 years before diagnosis. Comparable findings were confirmed in the subgroup of never and former smokers. On the other hand, previously reported results in the same population had suggested an increased risk for exposures closer in time to the year of diagnosis (i.e., 2000)¹⁶. We further confirmed this difference not only by comparing the results from the two years but also by considering the two exposures combined, mutually adjusting for one another.

The observed differences might be related to several factors. First, distinct methodologies were used to estimate exposure: FARM chemical transport model was used to estimate PM₁₀ in 1990²¹ while PM₁₀ estimates for the year 2000 were derived from a hybrid spatial-temporal model²⁷. Second, the granularity of 1990 exposure estimates was coarser than in 2000. For this reason, we rescaled the spatial resolution of exposure estimates in 2000 to a coarser definition (3 × 3 km) and obtained results that were comparable to those from the old analysis. Third, dissimilar levels in pollutant concentrations in the two time periods might also justify the observed difference in air-pollution related lung cancer risk, although the direction of the association is counterintuitive: if we compare PM₁₀ concentration levels, we observe that the average pollutant levels in 1990 were 46% higher than in 2000. This phenomenon is particularly evident in the Milan area, where PM₁₀ levels markedly decrease in the timespan we are considering. Fourth, we might look at our results within the greater framework of tumorigenesis, which was proposed to occur as a two-step process: i.e., the first one inducing mutations in healthy cells, and the second one triggering cancer development³². Our results might thus support a greater relevance of more recent exposures, especially when considering the two exposures combined. To this regard, it is interesting to note that a recent paper by Hill and colleagues provided combined results from a plethora of observational, in vivo, and in vitro studies to support the hypothesis that “environmental particulate matter measuring ≤ 2.5 µm […] promotes lung cancer by acting on cells that harbour pre-existing oncogenic mutations in healthy lung tissue”³³. In addition, in one of the studies they performed on about 250 female lung cancer cases, they observed an association with lung cancer risk when considering 3-year PM_2.5 cumulative exposure, while this association was not evident after 20-year exposure to the same pollutant.

Several other studies investigated the association between long-term air pollution exposure and lung cancer. Starting from the studies evaluated by the IARC, Hamra et al. conducted a large meta-analysis including 18 studies from different countries in Europe and North America. Data on PM₁₀ were available for nine studies (five in North America, three in Europe and one in New Zealand). Overall, a positive association was observed for PM₁₀ (RR: 1.08; 95% CI: 1.00-1.17), with mean levels ranging from 8.3 to 51 µg/m¹¹. Looking in more detail at the three studies conducted in Europe,

the Multicenter ESCAPE study investigated the effects of long-term PM₁₀ exposure on lung cancer incidence, using data from 14 cohorts. The enrollment years ranged from 1985 to 2005. Air pollution concentrations at the baseline residential addresses of the study subjects were back-extrapolated using land-use regression models applied to pollution data collected between 2008 and 2011 (i.e., after the study period). Back-extrapolated PM₁₀ levels ranged from about 5 to 90 µg/m³, with the highest levels estimated in Italy and Greece. The pooled analysis showed an association between the risk of lung cancer and increasing PM₁₀ concentrations (HR: 1.22; 95% CI: 1.03–1.45 per 10 µg/m³)³⁴.

The English study followed up a national cohort from 2003 to 2007 and observed positive associations for an interquartile range (IQR) increase in PM₁₀ (HR: 1.03; 95% CI: 0.98–1.08), using exposure data in 2002 (1–5 years before death) and adjusting for smoking status, cigarettes/day and income³⁵. PM₁₀, estimated from emission-based models, ranged from 12.6 to 29.8 µg/m³.

The mortality of a small cohort of women enrolled between 1985 and 1994 was examined through 2008 in Germany. PM₁₀ in the years 1985–1994 was derived from total suspended particles (TSP, applying a coefficient of 0.71) measured by fixed monitoring sites. An increase of 7 µg/m³ in PM₁₀ was associated with an increased HR of 1.84 (95%CI: 1.23–2.74), in models adjusted for education and smoking status³⁶.

Among the five US studies included in the meta-analysis by Hamra and colleagues, the Adventist Health Study on Smog (ASHMOG) followed up for cancer incidence (1977–1992) a large cohort of 6,338 non-smoking Californian adults. PM₁₀ levels were derived from TSP from 1972 to 1987 and subsequently directly measured through fixed-site monitoring stations. PM₁₀ ranged from 0 to 85 µg/m³ with mean values of 51.0 µg/m³. Incident lung cancer in males was positively associated with an IQR increase (24 µg/m³) in PM₁₀ mean concentrations (RR: 5.21, 95%CI: 1.94–13.99, adjusted by cigarette pack-years, education, and alcohol)³⁷.

A cohort study performed by the American Cancer Society within the Cancer Prevention Study (ACS-CPS-II) examined the association between air pollution and mortality among over 400,000 people enrolled in 1982–1998. Mean PM₁₀ concentration derived from fixed monitoring stations in the enrollment period was 28.8 µg/m³. Unfortunately, numerical results for PM₁₀ were not available, and a relative risk slightly greater than 1.00 in the period 1987–1996 (with 95%CI including the null value) could be derived only by looking at one of the paper figures³⁸.

The association between PM₁₀ exposure and cancer mortality (1985–2000) was investigated in about 40,000 workers employed in the U.S. trucking industry. PM₁₀ annual average exposures were determined for 1985 through 2000 from a model using spatial smoothing and geographical information system-based covariates. The average exposure during the study period was 26.8 µg/m³. A HR for lung cancer of 1.08 (95%CI: 0.91–1.30) was estimated per 10 µg/m³. No data on smoking were available³⁹.

The California teachers Study (TCS) involved a large cohort of women followed up for mortality (1997–2005)⁴⁰. Monthly average PM₁₀ concentrations were calculated from fixed-site monitors for the period 1996–2005. PM₁₀ estimates ranged from 9.19 to 82.64 µg/m³ (mean: 29.21). The HR per 10 µg/m³, adjusted for smoking status, pack-years, body mass index, alcohol, passive smoking, dietary fat, fiber and calories, was 0.93 (95%CI: 0.81–1.07).

The relationship between lung cancer incidence and long-term residential exposure to PM₁₀ was also evaluated within the Nurses’ Health Study, another large cohort of women followed up from 1994 to 2010⁴¹. PM₁₀ ranged from 3.17 to 74.79 µg/m³. Exposure was defined as the time-varying cumulative PM₁₀ average experienced in the previous 72-months. The HR per 10 µg/m³, adjusted for income, (active and passive) smoking and dietary variables, was 1.04 (95%CI: 0.95–1.14).

Finally, in the New Zealand Census-Mortality study, records from the 1996 national census were probabilistically linked to mortality data (1996–1999)⁴². Average PM₁₀ exposure at 1996 census were estimated using land use regression models and ranged from 0 to 30 µg/m³. The OR per 10 µg/m³ of PM₁₀, adjusted for ethnicity, social deprivation, income, education, and smoking status was 1.14 (95%CI: 1.05–1.23) among subjects who had not changed residence since 1991.

In summary, the abovementioned studies return a heterogeneous picture, where different time windows of exposure are investigated, concentration levels cover a quite wide range of values, and results are often characterized by some uncertainty, notwithstanding the overall interpretation of the findings that point towards an increasing trend in risk.

Following the IARC evaluation other studies were published.

The Nation-wide Dutch Environmental Longitudinal Study (DUELS) examined the mortality (2004–2011) of a cohort of more than 7 million subjects⁴³. Applying land-used regression models, annual mean PM₁₀ concentrations in 2001 were derived. The median PM₁₀ concentrations was 29 µg/m³. The HR, adjusted for age, sex, marital status, region of origin, household income, per 10 µg/m³ increase in PM₁₀ was 1.26 (95%CI: 1.21–1.30). No information on smoking and diet was available.

In a multi-cohort Swedish study, high-resolution dispersion models were used to estimate annual mean concentrations of PM₁₀ at individual address during each year of follow-up (1990–2011), and moving averages were calculated for the time-windows 1–5 years and 6–10 years preceding the outcomes of interest (i.e., natural and cause-specific deaths). No associations were observed between PM₁₀ exposure (range: 6.3–41.9 µg/m³) and lung cancer mortality, in models adjusted for sex, calendar year, sub-cohort, smoking status, alcohol consumption, physical activity, marital status, socioeconomic status, educational level, and occupation⁴⁴.

Moving to lung cancer incidence, several studies were conducted in South Korea. A large cohort study included more than 6.5 million subjects from the Korean National Health Insurance Service (NHIS) database for 2006–2007, who were linked to Korea Central Cancer Registry data to confirm lung cancer incidence for 2006–2013. Individual exposure to PM₁₀ was assessed as five-year average concentrations predicted at subjects’ district-specific home addresses for 2002–2007 via a geo-statistical model (mean: 55.8 µg/m³). An increased risk in lung cancer incidence (comparing the upper vs. the lower quartile of exposure) was estimated only for adenocarcinoma in current male smokers (HR: 1.14, 95%CI: 1.03–1.25, in models adjusted for age, gender, income, BMI, pack-year, history of COPD, interstitial lung disease, lung disease due to external factor, physical activity and alcohol consumption)⁴⁵. Another NHIS-based study was conducted in the Seoul metropolitan area, where about 83,000 people were followed up for 2007–2015. PM₁₀ exposure was estimated predicting district-specific annual-average concentrations (mean: 56.2 µg/m³). HRs for lung cancer incidence ranged from 1.05 (95%CI: 0.81–1.36) to 1.14 (95%CI: 0.84–1.54) considering either a 1-year or 5-year exposure average, respectively⁴⁶. A further update of this study was provided, taking also advantage of improved exposure assessment: about 2 million people without history of lung cancer in 2002–2006 were selected from the NHIS database. Lung cancer incident cases were identified up to 2016, and PM₁₀ concentration levels were estimated as 5-year averages at subjects’ residential addresses geocoded on a 100-m grid through a prediction model. PM₁₀ levels ranged from 28.6 to 100.7 µg/m³ and were not associated with increased lung cancer risk, regardless of the various adjustments applied, which included both individual and areal covariates⁴⁷.

The potential of the UK Biobank was exploited to investigate whether air pollution and genetic factors could jointly contribute to incident lung cancer. Almost 460,000 participants aged 40–69 years, recruited between 2006 and 2010, and without previous cancer at baseline, were included. PM₁₀ concentrations at residential address were estimated applying land-use regression models: estimates were available for years 2007 and 2010, and the average of these sets of values was used as mean exposure of the study participants (19.3 µg/m³). In the fully adjusted model (i.e., including age, sex, BMI, household income, education level, smoking status, and pack-years of smoking), a 10 µg/m³ increase in PM₁₀ concentration was associated with an HR of 1.53 (95%CI: 1.20–1.96). Interestingly, subjects with a high “genetic risk” (based on gene polymorphisms) had a higher risk of incident lung cancer (HR: 1.77, 95%CI: 1.50–2.10)⁴⁸. Another analysis on a smaller population selected from the UK biobank (N ≈ 367,000) tried to consider also the effect of lifestyle factors, and the association between PM₁₀ exposure and lung cancer risk was smaller and slightly more uncertain (HR: 1.05, 95%CI:0.99–1.10 per IQR increase in PM₁₀, in models adjusted for age, sex, race, educational level, household income, region, BMI, alcohol status, smoking status, physical activity, and diet)⁴⁹.

The role of PM₁₀ exposure on both mortality and hospitalization for lung cancer was investigated in a recent cohort study conducted in Guangzhou, China, where about 650,000 participants were recruited between 2009 and 2015, and followed up for an average of 8.2 years. Annual mean concentration of PM₁₀ (58.8 µg/m³) was estimated using a satellite-based random forest algorithm. Applying a causal inference model, for every 1 µg/m³ increase in PM₁₀ concentration the HR for mortality due to lung cancer was 1.032 (95%CI: 1.024–1.041), and that for hospitalization was 1.067 (95%CI: 1.020–1.115)⁵⁰.

The only case-control study that we know of specifically investigating PM₁₀ exposure and lung cancer was performed in South Korea⁵¹. Nine hundred and eight incident lung cancer cases were matched to population controls by sex, smoking status and age. PM₁₀ exposure assessment was done using land-use regression models in a 20-year period preceding subjects’ enrollment and complete residential histories since 1995. PM₁₀ levels ranged from 18.25 to 95.58 µg/m³. The OR adjusted for education, smoking status, education, exposure to occupational carcinogens, and fruit intake was 1.09 (95%CI: 0.96–1.23) per 10 µg/m³ of PM₁₀. Compared to the EAGLE study, this investigation is limited by the poor characterization of information regarding smoking habits (i.e., only smoking status was available).

In summary, the evidence trying to link PM₁₀ and lung cancer risk has been steadily growing and points towards the carcinogenic effects of this air pollutant. Nonetheless, uncertainties characterizing the various investigations must be acknowledged, probably related to the variability of the confounders considered in the analyses, the different methods applied for the exposure assessment, the wide range of the pollutant concentration values, and the different time windows investigated (very often close to the health event of interest).

To this regard, the role of exposure to air pollution 10 years before lung cancer diagnosis has been recently examined in a large Southern Californian cohort study including more than 45,000 participants aged 50 + years at enrollment. Lung cancer risk was estimated considering 5-year moving averages of ultrafine particles (UFP) exposure, lagged 10 years⁵². Interestingly, the authors found only a modest association between UFP exposure and lung cancer risk when considering the entire cohort (HR: 1.03, 95%CI: 0.99–1.08), while an increased risk was observed for adenocarcinoma when comparing most exposed vs. least exposed males (HR: 1.39, 95%CI: 1.05–1.85), and also in male participants who were oldest at the time of enrollment (HR: 1.13, 95%CI: 1.02–1.26). Notwithstanding the far smaller sample size of our study population and the difference in the considered pollutants (although combustion-derived ultrafine particles are numerous in urban PM₁₀⁵³), we tried to replicate the analyses by Jones and colleagues: however, no positive associations were observed neither when looking at adenocarcinoma in males, nor when further restricting our analysis to the same histological subtype among males aged 60 years or more at enrollment (results not shown). Possible explanations of these partially contrasting findings (apart from the obvious difference in sample size and pollutants considered among the two studies) might be the finer spatial resolution of exposure levels in the California study (LUR models to predict average levels at participant residential addresses) but also, on the other hand, the lack of control for important confounders such as occupational exposures, for which indeed we were able to control for.

Our study has several strengths. First, EAGLE is a population-based case-control study with a large number of enrolled subjects¹⁸. Many variables regarding residential history, smoking habits, and several others factors relevant for lung cancer were collected. This makes us confident of the quality of our data and allowed us to control for many potential confounders or risk factors that could bias the effect of air pollution in the analysis. Second, we assessed a past exposure to air pollution temporally located 12–15 years before the enrollment which, as far as we know, has rarely been evaluated. Third, we are dealing with exceptionally high values of air pollution concentrations: the mean PM₁₀ estimate of exposure of our study population in 1990 was 68 µg/m³, i.e., more than four times the air quality guideline level recommended by the World Health Organization (15 µg/m³)⁵⁴. Notably, we also had estimates of PM_2.5 values in 1990. We did not focus our analyses on PM_2.5 because our main aim was to make a comparison with our previous results regarding the year 2000, where only PM₁₀ estimates were available. However, PM_2.5 levels represented about 96% of PM₁₀ values in our dataset. As such, the annual average PM_2.5 level in our study population would roughly be 65 µg/m³, i.e., more than three times higher than the mean of the highest quartile of exposure in the study by Jones and colleagues (21.8 µg/m³). In any case, we replicated our main analysis considering PM_2.5 exposure and found comparable results (Table S9). Fourth, by comparing exposures in two time points, we tried to address the insidious topic of temporality in cancer development.

This study has also limitations. First, the spatial distribution of exposure estimates in 1990 is very different from that in 2000. This is particularly evident in Milan, which has much higher PM₁₀ values as compared with other areas. A second limitation regards the low spatial resolution of our exposure data. The model applied to estimate 1990 PM₁₀ concentration levels used a 3 × 3 km grid, thus implying that all subjects localized within the same cell were assigned the exact same exposure value, possibly introducing a nondifferential misclassification of exposure. To this regard, we must also consider that about 70% of our study population was resident within the metropolitan city of Milan. These subjects all fall within four cells only: as such, in our analysis for the year 1990 most subjects share the same exposure. However, the sensitivity analysis where we rescaled exposure estimates for the year 2000 to the 3 × 3 km grid basically reconfirmed previous results obtained with exposure estimates at a much finer spatial resolution, suggesting that exposure granularity is not the main driver of the different findings. Finally, as already discussed, we are comparing different methodologies to estimate exposures in the two time points, which could affect results with their own uncertainties⁵⁵.

Introduction

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin’s lymphoma (NHL) and is associated with poor outcomes in patients who were refractory to or relapsed after first-line R-CHOP therapy. Glofitamab, a CD20×CD3 T-cell-engaging bispecific monoclonal antibody, has emerged as a promising therapeutic option for adult patients with relapsed or refractory B-cell NHL, particularly those who have received at least two prior lines of systemic therapy.

Antisynthetase syndrome (ASyS) is a distinct clinical entity within the spectrum of idiopathic inflammatory myopathies (IIM) and is considered a protective factor against hematological malignancies, rendering the coexistence of DLBCL and ASyS exceptionally rare. However, this comorbidity presents unique therapeutic challenges due to potential bidirectional influences between the two diseases. On the one hand, the immunosuppressive agents required to manage ASyS, such as glucocorticoids, calcineurin inhibitors, or JAK inhibitors, may impair anti-lymphoma immune responses and reduce the efficacy of lymphoma-directed therapies. On the other hand, cytotoxic or immunomodulatory agents used to treat DLBCL may exacerbate or unmask autoimmune manifestations. Furthermore, the risk of treatment-related infections is increased due to overlapping immunosuppressive mechanisms.

The advent of chimeric antigen receptor T (CAR-T) cell therapy and T-cell-engaging bispecific antibodies has stimulated interest in their potential application beyond hematologic malignancies, including autoimmune diseases (AID). Several case reports have demonstrated the successful use of CD19 CAR-T cell therapy in refractory AID, such as rheumatoid arthritis, systemic lupus erythematosus (SLE), and ASyS. However, to date, there have been no published reports describing the use of glofitamab in patients with autoimmune diseases. Here, we report a case of DLBCL in a patient with a long-standing history of ASyS. The patient achieved complete remission (CR) of lymphoma with third-line glofitamab therapy after failing multiple prior treatments. Remarkably, her ASyS symptoms, which had been refractory to numerous immunosuppressive and targeted agents, also resolved after glofitamab therapy.

Case Presentation

A 46-year-old female presented with a 14-year history of recurrent rashes and myalgia, and a one-month history of a pulmonary mass. In 2009, she developed violaceous rashes on her anterior chest and rough, thickened skin on her palms, accompanied by bilateral knee joint pain. Laboratory tests revealed elevated muscle enzymes, with creatine kinase (CK) levels at 330 IU/L. She was diagnosed with connective tissue disease and responded to glucocorticoid treatment. In 2018, her rash worsened, accompanied by recurrent fever, muscle soreness, and lower limb weakness. Electromyography indicated myogenic damage, while a muscle biopsy revealed degeneration, necrosis, and regeneration of muscle fibers, along with focal perivascular lymphocyte infiltration, consistent with inflammatory myopathy. Chest computed tomography (CT) revealed interstitial lung disease. Antibody screening showed a positive antinuclear antibody (ANA) titer of 1:0000 with a cytoplasmic granular pattern, anti-Ro-52 (++), and anti-Jo-1 (+++), leading to a diagnosis of ASyS. Between 2019 and 2023, she was treated with glucocorticoids, various immunosuppressants (cyclosporine, methotrexate, hydroxychloroquine), and targeted therapies (tofacitinib, baricitinib) without significant improvement.

In February 2023, a follow-up chest CT revealed two soft tissue masses adjacent to the pleura, one in the middle lobe and one in the lower lobe of the right lung (Figure 1A–D). CT-guided lung biopsy confirmed DLBCL, non-germinal center type. PET/CT showed increased metabolic activity in a subpleural mass in the right lower lobe (2.6×3.9 cm, SUVmax 15.14), another subpleural mass in the right middle lobe (5.3×3.4 cm, SUVmax 7.3), bilateral axillary lymph nodes (0.5–1.1 cm, SUVmax 4.1), and multiple subcutaneous nodules. The patient was diagnosed with DLBCL, Ann Arbor stage IV, with an International Prognostic Index score of 4. She received eight cycles of R-CHOP (rituximab combined with cyclophosphamide, doxorubicin, vincristine, and, prednisone) therapy as first-line treatment. End-of-treatment PET/CT showed a slight reduction in the size and metabolic activity of the right middle lobe mass, but increased metabolic activity in the right lower lobe mass, indicating disease progression (Figure 1E–H). A repeat biopsy of the right lung mass reconfirmed the diagnosis of DLBCL. During this time, she continued to experience recurrent fever, rashes, and muscle weakness.

Figure 1 Computed tomography (CT) and positron emission computed tomography (PET) of this patient. (A–C) CT finding of the pulmonary lesions of on Feb 2023; (B–D) PET showing increased metabolic activity of the pulmonary lesions of on Feb 2023; (E–H) The pulmonary lesions on CT scan and PET/CT after 8 cycles of R-CHOP therapy; (I–L) Significant reduction of the pulmonary lesions with decreased metabolic activity after 4 cycles of glofitamab therapy.

The patient was subsequently treated with second-line polatuzumab vedotin combined with lenalidomide, but no remission was achieved. Third-line therapy with the CD20×CD3 bispecific antibody glofitamab was initiated. After four cycles, follow-up PET/CT revealed decreased metabolic activity and reduced size of the right middle and lower lobe masses (SUVmax 1.8), indicating complete metabolic remission (Figure 1I–L). During glofitamab treatment, the patient experienced grade 1 cytokine release syndrome (CRS), grade 2 leukopenia, and grade 2 thrombocytopenia, with no serious infections. In addition to the successful lymphoma treatment, the patient showed significant improvement in fever and rashes. As of October 2024, she has completed eight cycles of glofitamab, with lymphoma remaining in sustained CR. Following treatment, the patient experienced significant improvement in her dermatomyositis symptoms. Objective indicators of ASyS also improved, including normalization of CK levels and conversion to negative ANA and anti-Jo-1 antibody status. Figures 2A and B depict the dynamic changes in white blood cell count, hemoglobin, and platelet levels during glofitamab treatment. Figure 2C presents a timeline summarizing the patient’s clinical course.

Figure 2 A timeline illustrating the clinical course and treatment of this case. (A) Changes of the white blood cell counts and lymphocyte counts during glofitamab treatment; (B) Changes of the hemoglobin and platelet counts during glofitamab treatment. (C) A timeline illustrating the clinical course and treatment of this case.

Discussion

Glofitamab is a CD20×CD3 T-cell-engaging bispecific monoclonal antibody under development for the treatment of B-cell NHLs.¹ In a Phase II trial evaluating glofitamab in patients with relapsed or refractory DLBCL who had received at least two prior lines of therapy, encouraging results were reported, with a CR rate of 39% and a 12-month progression-free survival rate of 37%.² Glofitamab has also demonstrated efficacy in patients with DLBCL relapsing after CAR-T cell therapy, inducing the expansion of residual CAR-T cells.³ In this case, the patient, who was primarily refractory to R-CHOP therapy and did not respond to polatuzumab vedotin combined with lenalidomide, received glofitamab as third-line treatment and achieved CR.

The patient’s medical history included a prior diagnosis of ASyS, preceding the onset of lymphoma. ASyS is a distinct entity within the spectrum of IIM.⁴ Patients with IIM are recognized to have an increased risk of malignancy, with symptoms often preceding tumor onset. However, within the IIM spectrum, ASyS has been reported as a protective factor against hematological malignancies.⁵ In this patient, typical dermatomyositis symptoms began 14 years before the lymphoma diagnosis. Given the delayed onset of lymphoma and the relatively low malignancy risk associated with ASyS, ASyS was not considered a paraneoplastic syndrome in this case. Interestingly, the patient’s ASyS symptoms resolved following glofitamab treatment, coinciding with the lymphoma response. Prior treatment for ASyS with various immunosuppressive agents, JAK inhibitors, and rituximab yielded unsatisfactory results. Unexpectedly, her ASyS symptoms responded well to glofitamab, with ANA and anti-Jo-1 antibodies becoming negative.

With the advent of CAR-T cell therapy, which has shown promising outcomes in B-cell NHLs, researchers have explored its potential application in AID, including rheumatoid arthritis, SLE, and multiple sclerosis. Emerging evidence suggests that CD19 CAR-T cell therapy may effectively manage refractory AID. Preliminary studies have indicated that CD19 CAR-T cell therapy is a safe and effective therapeutic option in SLE.^6,7 Case reports have also described successful use of CD19 CAR-T cell therapy in refractory ASyS.⁸ Similar to CAR-T cell therapy, T-cell–engaging bispecific antibodies recruit T cells to mediate B-cell cytotoxicity, presenting a potential alternative strategy for treating refractory AID. Schreiber et al recently reported the effective use of a BCMA-targeting bispecific T-cell–engaging antibody in a patient with treatment-refractory LRP4-positive myasthenia gravis.⁹ To the best of our knowledge, this is the first report of successful ASyS treatment with a CD20×CD3 T-cell–engaging bispecific antibody, glofitamab.

T-cell engagers (TCEs) may offer certain advantages over CAR-T cell therapy in AID treatment. First, TCEs enable precise dosing and treatment duration, offering greater control compared to CAR-T cell therapy. Additionally, TCE therapy does not require lymphodepletion before infusion, potentially lowering the risk of infection. Furthermore, TCEs do not require the lag time associated with leukapheresis and CAR-T cell engineering, allowing for more rapid treatment initiation. These features suggest that TCEs may address some limitations of CAR-T cell therapy in AID management. However, the concurrent use of immunosuppressants, such as mycophenolate mofetil, may impair T-cell function and reduce the efficacy of both CAR-T cell and TCE therapies.¹⁰ To mitigate this, withholding immunosuppressants for a period to allow T-cell recovery may be considered in both treatment approaches.¹¹

While this case highlights the potential immunomodulatory effects of T-cell-engaging bispecific antibodies, the use of TCEs in AIDs also raises important considerations regarding the balance between efficacy and safety. Several risks must be considered. First, TCEs can induce CRS and immune effector cell–associated neurotoxicity syndrome, which may be particularly hazardous in patients with underlying systemic inflammation or end-organ dysfunction. Second, excessive or prolonged B-cell depletion may increase susceptibility to infections and impair vaccine responses. Careful patient selection, vigilant monitoring, and a deeper understanding of the immune contexture in AIDs are therefore essential when considering TCEs in this setting.

In conclusion, this single-case observation raises the possibility that T-cell–engaging therapies may offer benefit in refractory autoimmune diseases (AIDs), particularly in immunologically complex settings. However, given the inherent limitations of an isolated case, no definitive conclusions can be drawn. Ultimately, prospective clinical studies will be essential to evaluate the safety, efficacy, and broader applicability of TCEs in patients with AID.

Data Sharing Statement

Data supporting the findings of this study are available upon reasonable request from either of the corresponding authors, Dr. Yan Zhang or Dr. Wei Zhang.

Ethics and Consent

The patient had given written informed consent for the publication of any potentially identifiable images or data included in this article. Institutional approval from Peking Union Medical College Hospital review board was not required to publish the case details.

Disclosure

The authors report no conflicts of interest in this work.

References

1. Bacac M, Colombetti S, Herter S, et al. CD20-TCB with obinutuzumab pretreatment as next-generation treatment of hematologic malignancies. Clin Cancer Res. 2018;24(19):4785–4797. doi:10.1158/1078-0432.CCR-18-0455

2. Dickinson MJ, Carlo-Stella C, Morschhauser F, et al. Glofitamab for relapsed or refractory diffuse large B-cell lymphoma. N Engl J Med. 2022;387(24):2220–2231. doi:10.1056/NEJMoa2206913

3. Rentsch V, Seipel K, Banz Y, et al. Glofitamab treatment in relapsed or refractory DLBCL after CAR T-cell therapy. Cancers. 2022;14(10):2516. doi:10.3390/cancers14102516

4. Tiniakou E, Mammen AL. Idiopathic inflammatory myopathies and malignancy: a comprehensive review. Clin Rev Allergy Immunol. 2017;52(1):20–33. doi:10.1007/s12016-015-8511-x

5. Mammen AL. Paraneoplastic myopathies. Handb Clin Neurol. 2024;200:327–332.

6. Mackensen A, Müller F, Mougiakakos D, et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nat Med. 2022;28(10):2124–2132. doi:10.1038/s41591-022-02017-5

7. Li M, Zhang Y, Jiang N, et al. Anti-CD19 CAR T cells in refractory immune thrombocytopenia of SLE. N Engl J Med. 2024;391(4):376–378. doi:10.1056/NEJMc2403743

8. Müller F, Boeltz S, Knitza J, et al. CD19-targeted CAR T cells in refractory antisynthetase syndrome. Lancet. 2023;401(10379):815–818. doi:10.1016/S0140-6736(23)00023-5

9. Schreiber S, Al-Dubai M, Vielhaber S, et al. Effective use of BCMA-targeting bispecific T cell-engaging antibody in treatment-refractory LRP4+ myasthenia gravis. Mol Ther. 2025;S1525-0016(25):00480.

10. Nakamura M, Ogawa N, Shalabi A, et al. Positive effect on T-cell regulatory apoptosis by mycophenolate mofetil. Clin Transplant. 2001;15(Suppl 6):36–40. doi:10.1034/j.1399-0012.2001.00006.x

11. Mamlouk O, Nair R, Iyer SP, et al. Safety of CAR T-cell therapy in kidney transplant recipients. Blood. 2021;137(18):2558–2562. doi:10.1182/blood.2020008759