Since the 1970s, total body irradiation (TBI) has played a central role in conditioning regimens prior to hematopoietic stem cell transplant (HSCT) (1, 2). The goal of TBI is to ablate any remaining malignant cells and/or to reduce risk of transplant rejection by modulating the immune system. Historically, TBI has been administered using 2-dimensional planning, which limits sparing of normal organs, and leads to significant possible adverse toxicity. Indeed, typically attempts are only made to spare the lung by using lung blocks (3). However, even that methodology is suboptimal. In an analysis of a Children’s Oncology Group (COG) trial, lung dose <8 Gy was associated with improved overall survival, but can be difficult to achieve with conventional TBI (4). Furthermore, the analysis found that lung shielding during TBI is not standardized, with lung doses ranging from 50% to the full TBI dose, justifying the need for more modern radiation techniques to reduce dose to organs at risk (OARs).

One such approach has been to use a more modern 3-dimensional radiation technique in intensity modulated radiation therapy (IMRT) to deliver TBI, which has the capability of delivering a more sculpted and conformal dose distribution. The utility of such an approach has previously been described, with tomotherapy (Tomo IMRT) reducing the mean lung dose from 8-9 Gy to 5-6 Gy, and reduced dose correlating with significantly less pulmonary complications in patients treated with total marrow and lymphoid radiation (TMLI) (5, 6). Multiple trials have reported on using IMRT TBI, using either a Tomo IMRT (7) (8) or volumetric modulated arc therapy (VMAT) technique (9, 10). However, given the recent publication of IMRT TBI studies, there is a paucity of long-term safety and efficacy data.

Recent retrospective studies have offered some insight into the long-term outcomes of IMRT TBI. One Russian study of 220 pediatric patients who received TBI- and chemo-conditioning for allogeneic HSCT with TCRαβ/CD19 depletion reported an overall survival (OS) of 63% and event-free survival (EFS) of 58% at a median follow-up of 2.8 years (11). Another retrospective of 44 pediatric and adult patients at UT Southwestern who received low- and standard-dose VMAT-TBI found that median time to relapse was 11.3 months at a follow up of 2.2 years, and reported that acute and chronic GVHD occurred in 59% and 39% of patients, respectively (12). Most recently, researchers at Stanford reported on a group of 38 pediatric and adolescent/young adult (AYA) patients who received VMAT-TBI at both myeloablative and non-myeloablative doses. At a median follow-up of 8.7 months, they reported an OS of 89.5% and relapse-free survival (RFS) of 94.7% for all patients (13). While these studies emphasize the safety and efficacy of IMRT TBI, their median follow-up periods do not exceed three years, limiting our understanding of the long-term effects of IMRT TBI.

Since 2010, our institution has been performing Tomo IMRT TBI off-protocol for patients who cannot tolerate standing for treatment or who have had prior radiation. Therefore, we performed a retrospective matched-pair analysis of the treatment and outcomes of these patients compared to patients who received IMRT TBI. Herein, we report on their long-term follow-up, including oncologic outcomes and associated adverse effects.

Advances in radiation technology have led to implementation of more advanced radiation techniques such as IMRT into the transplant conditioning regimen. Several studied have reported on the procedure for performing IMRT TBI, including initial clinical outcomes (11–13). However, among these studies, the highest median follow-up was 2.8 years, with a maximum follow up of 6.4 years (11). Therefore, there is a need for evaluation of long-term clinical outcomes and toxicities. To our knowledge, our study offers the longest follow-up of patients treated with IMRT TBI, with a median follow-up of 5.1 years for living patients and a maximum follow-up of 10.7 years. In our study, we observe favorable long-term outcomes and rates of toxicity in our IMRT TBI cohort, which should support the continued investigation of IMRT TBI as a conditioning regimen for patients undergoing hematopoietic cell transplantation.

The improved ability of IMRT TBI to spare critical organs may help mitigate toxicities. Although there were variations in prescription dose, all patients in our cohort achieved the recommended mean lung doses of <800 cGy, which has been shown to be difficult to achieve with conventional blocks (4, 15). The value of this is illustrated by the lack of pneumonitis in our IMRT TBI cohort, while there were several cases in the standard TBI cohort. Similarly, almost all patients achieved a mean kidney dose of less than <12 Gy, with no nephrotoxicity events, though this was not unexpected, with low rates at baseline and no nephrotoxicity in the standard cohort either. Overall, our dosimetry is similar to results reported by Marquez et al., who achieved mean lung dose of 7.3 Gy and mean kidney dose of 71.4% of prescription dose, and also experienced no pneumonitis or nephritis (13). In fact, our mean doses of 6.2 Gy and 5.7 Gy respectively reflect potential to further limit exposure of critical organs, which may improve transplant-related morbidity and mortality (16). Our organ sparing still yielded high PTV coverage, with a median of 85% achieving prescription dose, and led to favorable oncologic outcomes.

Importantly, the 5-yr survival outcomes in our study does not appear to be negatively impacted by organ sparing. Indeed, we observed no recurrences nor deaths beyond two years and survival was not significantly different from the matched standard TBI cohort. Although there were two cases of transplant-related mortality within a month of transplant in the IMRT cohort, these deaths were attributed to neutropenic fever/sepsis and seizures, which are almost certainly unrelated to the TBI technique. Conditioning regimen is a critical contributor to risk of early relapse following transplant, so deintensifications such as IMRT TBI should be assessed critically (17). In total, there were three relapses, all of which were limited to the bone marrow and got the full radiation dose. Therefore, these relapses are almost certainly unrelated to TBI technique. Further, although heterogeneity across published and our own IMRT TBI cohorts limits cross study comparisons of survival, our 2-year outcomes are comparable to other reported series, where 2-yr OS varied from 63-79% and EFS ranged from 58-71% (11, 12).

Although other organs, such as the lens and oral cavity were only spared in select cases, the minimum doses to those structures were 3.6% and 35.4% or prescription dose, respectively. Such doses would not be feasible with traditional TBI. With limited patient numbers, we cannot be certain of the impact this sparing might have on long-term toxicity such as cataract formation (notably no patients in our IMRT cohort reported cataract development over the course of follow-up), though analyses of dose response for radiation cataractogenesis support this hypothesis (18). Although there were late toxicities observed in the standard TBI cohort, due to retrospective nature, low event rates, and small sample sizes, it is not possible to make conclusions on whether IMRT permits lower toxicity rates, though with organ sparing it is plausible and has been seen in related work (6).

Although currently not part of standard IMRT-TBI protocols, the ability to selectively target radiation dose and spare organs such as the lenses or oral cavity as part of the conditioning regimen allows for potential dose escalation of higher-risk areas while still sparing critical organs. This framework has provided the basis for similar work in the form of total marrow and lymphoid irradiation (TMLI), which had initially been used exclusively for high-risk patient populations who otherwise would not be transplant candidates (16, 19–21), but is now proceeding forward with trials in standard risk disease. Analysis of patients undergoing TMLI suggests organ sparing can lead to decreased toxicity in the form of lower rates of pneumonitis, nephrotoxicity, hypothyroidism, and cataract formation (6). Future studies involving IMRT TBI and TMLI will continue to seek to shape dose to optimize achieving disease control while limiting toxicities.

The incidence of GVHD, particularly in the standard TBI cohort, is relatively high compared to other more modern reports, though the etiology of this difference is unclear, given standard GVHD prophylaxis was administered and the vast majority of patients had matched donors (22, 23). Further, our institutional GVHD prophylaxis protocols are consistent with current recommended treatment standards. Overall, the rates of GVHD in our IMRT TBI cohort were similar to rates published by Zhang-Velten et al. (12) who also used IMRT TBI, as well as other non-IMRT TBI based regimens (24–27). There was also no difference in Grade III-IV aGVHD nor cGVHD in our standard TBI cohort compared to the IMRT TBI cohort. Interestingly, the IMRT TBI cohort did have significantly less grade II-IV aGVHD than the standard TBI cohort. The reason for this difference is not clear, however, there is an association between intensity of conditioning regimen and incidence of grade II-IV aGVHD (28). While the difference could be driven by differences in GVHD prophylaxis, patient population (though there were no significant differences in patient characteristics) or chance, it is possible that increased dose homogeneity and decreased hotspots in organs, such as the gastrointestinal tract, could be a cause (29). This would be an artifact caused by IMRT optimizations, given only lungs and kidney were contoured and intentionally spared. Interestingly, post-transplant cyclophosphamide may decrease aGVHD, though only patients in the standard TBI cohort received it (30, 31). Another possibility is contrast in dose rate between the techniques and associated differences in normal tissue damage from radiation (32). However, helical tomotherapy is associated with a dose rate much higher than standard TBI (33, 34), which one would expect to lead to greater toxicity, although this has not occurred in initial studies (6, 35, 36). Clearly, a better understanding of the relationship between radiation technique, dose rate, and transplant related morbidity will be necessary. Preclinical studies suggest aGVHD associated with radiation exposure level to associated organs, which can be attenuated using TMI (37–39). Future larger studies of IMRT TBI will optimally shed further light on whether technique might influence aGVHD.

Our study is limited by several features including small sample size and retrospective nature. Our small sample size also had significant heterogeneity, including multiple different hematologic malignancies and disease statuses at the time of transplant. This limits comparison of oncologic outcomes to other cohorts. We did seek to ameliorate this limitation by performing a matched pair analysis with patients receiving conventional TBI, which did not show any signal of any detriment to oncologic outcomes. The limitation also applies to assessment of late toxicity. Due to small patient numbers and the relative rarity of late toxicity events even in the general TBI population, it is difficult to draw definitive conclusions. This limitation is amplified by the retrospective nature of our study, since toxicities were not prospectively collected, and therefore had to be identified via chart review, which might miss some toxicities and generally precludes grading of toxicities. This also limits characterization of other non-malignant late effects and health-related quality of life. Nonetheless, the absence of significant late toxicities in our IMRT TBI cohort is promising. Despite these limitations, our study adds to the literature on IMRT TBI by providing some insight into long-term follow-up.

IMRT TBI appears to lead to favorable long-term outcomes, with a potential for improved long-term toxicity profile compared to conventional TBI associated with better organ sparing. Based on the results of our initial work, we have implemented a multi-institution pilot study (NCT04281199) of TBI using tomotherapy or VMAT-based IMRT in order to achieve new lung dose guidelines (4). We have also initiated a protocol for the treatment of scleroderma following positive results from the SCOT trial (40). IMRT TBI warrants further investigation as part of larger prospective trials.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by City of Hope National Medical Center Institutional Review Board. Written informed consent from the participants’ legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements.

CL and CIH performed data collection, data analysis, and manuscript preparation. DY participated in the design of this study and performed data collection and data analysis. SD participated in the design of this study, creation of methods for this study, and data analysis, and performed critical review of the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.