Development and validation of the lung donor (LUNDON) acceptability score for pulmonary transplantation

Developing approaches to incorporate donor-lung computed tomography images into machine learning models to predict severe primary graft dysfunction after lung transplantation

Primary graft dysfunction (PGD) is a common complication after lung transplantation associated with poor outcomes. Although risk factors have been identified, the complex interactions between clinical variables affecting PGD risk are not well understood, which can complicate decisions about donor-lung acceptance. Previously, we developed a machine learning model to predict grade 3 PGD using donor and recipient electronic health record data, but it lacked granular information from donor-lung computed tomography (CT) scans, which are routinely assessed during offer review. In this study, we used a gated approach to determine optimal methods for analyzing donor-lung CT scans among patients receiving first-time, bilateral lung transplants at a single center over 10 years. We assessed 4 computer vision approaches and fused the best with electronic health record data at 3 points in the machine learning process. A total of 160 patients had donor-lung CT scans for analysis. The best imaging-only approach employed a 3D ResNet model, yielding median (interquartile range) areas under the receiver operating characteristic and precision-recall curves of 0.63 (0.49-0.72) and 0.48 (0.35-0.6), respectively. Combining imaging with clinical data using late fusion provided the highest performance, with median areas under the receiver operating characteristic and precision-recall curves of 0.74 (0.59-0.85) and 0.61 (0.47-0.72), respectively.

Use of a novel donor lung scoring system as a tool for increasing lung recovery for transplantation

BACKGROUND

There is a critical shortage of donor lungs for transplantation. We previously developed a parsimonious, highly discriminatory 9-variable Lung Donor (LUNDON) acceptability score. Here we assessed the utility of this score as a tool for improving lung recovery rates for transplantation.

METHODS

We examined all brain-dead donors between 2014 and 2020 from 3 US organ procurement organizations and validated the score's predictive performance. We examined the trajectory of donors with low (<40) and high (>60) initial LUNDON scores, their corresponding lung recovery rates, factors contributing to score improvement using multivariable regression models, and 1-year post-transplant recipient survival.

RESULTS

Overall lung recovery was 32.4% (1410 of 4351). Validation of the LUNDON score in our cohort revealed a C statistic of 0.904 and required intercept calibration. Low initial LUNDON donors that improved to a high final score had an increase in lung recovery rate from 29.3% (1100 of 3765) to 86.8% (441 of 508), associated with lower body mass index, management in a specialized donor care facility (SDCF), and more bronchoscopies. Donors with high initial and final LUNDON scores had a lung recovery rate of 85.2% (98 of 115), associated with shorter length of hospital stay. One-year survival was similar in recipients of low-to-high versus high-to-high LUNDON score donors (0.89 vs 0.84; P = .2).

CONCLUSIONS

The LUNDON score performs well as a predictor of lung recovery in a contemporary cohort but may require organ procurement organization-specific calibration. SDCF care, increasing use of bronchoscopy, and decreasing the time from brain death to organ procurement may improve lung utilization. The LUNDON score can be used to guide donor management to expand the donor pool.

The perspective for next-generation lung replacement therapies: functional whole lung generation by blastocyst complementation

PURPOSE OF REVIEW

Blastocyst complementation represents a promising frontier in next-generation lung replacement therapies. This review aims to elucidate the future prospects of lung blastocyst complementation within clinical settings, summarizing the latest studies on generating functional lungs through this technique. It also explores and discusses host animal selection relevant to interspecific chimera formation, a challenge integral to creating functional human lungs via blastocyst complementation.

RECENT FINDINGS

Various gene mutations have been utilized to create vacant lung niches, enhancing the efficacy of donor cell contribution to the complemented lungs in rodent models. By controlling the lineage to induce gene mutations, chimerism in both the lung epithelium and mesenchyme has been improved. Interspecific blastocyst complementation underscores the complexity of developmental programs across species, with several genes identified that enhance chimera formation between humans and other mammals.

SUMMARY

While functional lungs have been generated via intraspecies blastocyst complementation, the generation of functional interspecific lungs remains unrealized. Addressing the challenges of controlling the host lung niche and selecting host animals relevant to interspecific barriers between donor human and host cells is critical to enabling the generation of functional humanized or entire human lungs in large animals.

Validation of a novel donor lung scoring system based on the updated lung Composite Allocation Score

Lung transplantation (LTx) continues to have lower rates of long-term graft survival compared with other organs. Additionally, lung utilization rates from brain-dead donors remain substantially lower compared with other solid organs, despite a growing need for LTx and the significant risk of waitlist mortality. This study aims to examine the effects of using a combination of the recently described novel lung donor (LUNDON) acceptability score and the newly adopted recipient lung Composite Allocation Score (CAS) to guide transplantation. We performed a review of nearly 18 000 adult primary lung transplants from 2015-2022 across the US with retroactive calculations of the CAS value. The medium-CAS group (29.6-34.5) had superior 1-year posttransplant survival. Importantly, the combination of high-CAS (> 34.5) recipients with low LUNDON score (≤ 40) donors had the worst survival at 1 year compared with any other combination. Additionally, we constructed a model that predicts 1-year and 3-year survival using the LUNDON acceptability score and CAS values. These results suggest that caution should be exercised when using marginally acceptable donor lungs in high-priority recipients. The use of the LUNDON score with CAS value can potentially guide clinical decision-making for optimal donor-recipient matches for LTx.

Lung Donation and Transplant Recipient Outcomes at Independent vs Hospital-Based Donor Care Units

Importance

Centralizing deceased organ donor management and organ recovery into donor care units (DCUs) may mitigate the critical organ shortage by positively impacting donation and recipient outcomes.

Objective

To compare donation and lung transplant outcomes between 2 common DCU models: independent (outside of acute-care hospitals) and hospital-based.

Design, Setting, and Participants

This is a retrospective cohort study of Organ Procurement and Transplantation Network deceased donor registry and lung transplant recipient files from 21 US donor service areas with an operating DCU. Characteristics and lung donation rates among deceased donors cared for in independent vs hospital-based DCUs were compared. Eligible participants included deceased organ donors (aged 16 years and older) after brain death, who underwent organ recovery procedures between April 26, 2017, and June 30, 2022, and patients who received lung transplants from those donors. Data analysis was conducted from May 2023 to March 2024.

Exposure

Organ recovery in an independent DCU (vs hospital-based DCU).

Main Outcome and Measures

The primary outcome was duration of transplanted lung survival (through December 31, 2023) among recipients of lung(s) transplanted from cohort donors. A Cox proportional hazards model stratified by transplant year and program, adjusting for donor and recipient characteristics was used to compare graft survival.

Results

Of 10 856 donors in the starting sample (mean [SD] age, 42.8 [15.2] years; 6625 male [61.0%] and 4231 female [39.0%]), 5149 (primary comparison group) underwent recovery procedures in DCUs including 1466 (28.4%) in 11 hospital-based DCUs and 3683 (71.5%) in 10 independent DCUs. Unadjusted lung donation rates were higher in DCUs than local hospitals, but lower in hospital-based vs independent DCUs (418 donors [28.5%] vs 1233 donors [33.5%]; P < .001). Among 1657 transplant recipients, 1250 (74.5%) received lung(s) from independent DCUs. Median (range) duration of follow-up after transplant was 734 (0-2292) days. Grafts recovered from independent DCUs had shorter restricted mean (SE) survival times than grafts from hospital-based DCUs (1548 [27] days vs 1665 [50] days; P = .04). After adjustment, graft failure remained higher among lungs recovered from independent DCUs than hospital-based DCUs (hazard ratio, 1.85; 95% CI, 1.28-2.65).

Conclusions and Relevance

In this retrospective analysis of national donor and transplant recipient data, although lung donation rates were higher from deceased organ donors after brain death cared for in independent DCUs, lungs recovered from donors in hospital-based DCUs survived longer. These findings suggest that further work is necessary to understand which factors (eg, donor transfer, management, or lung evaluation and acceptance practices) differ between DCU models and may contribute to these differences.