Researchers used inverse probability weighting regression adjustment to reduce confounding by measured baseline variables. In that adjusted model, subcutaneous administration was associated with higher odds of total pathological complete response, with an odds ratio of 1.79, a 95% confidence interval of 1.02 to 3.14 and a P value of 0.044. The abstract states that the direction of the association was consistent across complementary analyses, but the retrospective design means the results do not establish that the administration route caused the difference in response.

The investigators also examined whether baseline body weight modified the association between route of administration and total pathological complete response. Body weight was comparable between the groups, and the analysis found no significant interaction with continuous body weight after adjustment, with an interaction P value of 0.961. As a secondary radiological endpoint, serial ultrasound measurements showed broadly similar tumor and lymph-node size trajectories at the assessed treatment time points and did not discriminate the difference in final pathological response.

The findings are limited by the study’s single-center, retrospective nature and the imbalance between treatment groups, as well as its reliance on adjustment for measured baseline variables. The available abstract also identifies a need for prospective multicenter validation and longer oncological follow-up. It notes that randomized trials have demonstrated non-inferiority of subcutaneous versus intravenous trastuzumab, but those trial findings do not remove the uncertainties surrounding the path-completion association observed in this routine-practice cohort.

Subject of Research: Association between trastuzumab administration route and pathological complete response in HER2-positive breast cancer

Article Title: Association of neoadjuvant subcutaneous trastuzumab with higher pathological complete response in HER2-positive breast cancer: a real-world doubly robust study

Article References: Zhang, R., Liu, J., Wang, Z., Liu, Y., Shi, B., Zhang, C., Li, X., Feng, K., Li, X., Zheng, S., & Zhang, H. (2026). Association of neoadjuvant subcutaneous trastuzumab with higher pathological complete response in HER2-positive breast cancer: a real-world doubly robust study. BMC Cancer. https://doi.org/10.1186/s12885-026-17025-8

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17025-8

Keywords: HER2-positive breast cancer, subcutaneous trastuzumab, intravenous trastuzumab, pathological complete response, neoadjuvant therapy, retrospective cohort, inverse probability weighting, serial ultrasound, body-weight interaction, Association, neoadjuvant, subcutaneous

In this surgery population, 24 patients, or 77.4%, achieved a major pathological response, while 13, or 41.9%, achieved a pathological complete response. All 31 operations resulted in an R0 resection. Because the trial had no comparison group, the results demonstrate outcomes observed with the regimen but cannot determine how much, if any, were attributable specifically to toripalimab.

Before surgery, patients received radiotherapy delivered as 50 Gy/45 Gy/25 fractions, five cycles of concurrent weekly chemotherapy and four cycles of toripalimab, a PD-1 inhibitor. After radical surgery, patients received adjuvant chemoimmunotherapy. The primary endpoint was the major pathological response rate.

Secondary endpoints included pathological complete response, R0 resection, disease-free survival, overall survival, event-free survival and safety. The investigators also conducted exploratory analyses of immune correlates, including tertiary lymphoid structures, or TLS. Safety outcomes were limited in the reported cohort.

The investigators recorded one case of treatment-related toxicity of grade 4 or higher and one case of a surgical complication of grade 3 or higher. During a median follow-up of 54.4 months for overall survival and 51.6 months for disease-free survival, two- and four-year overall survival in the surgery cohort were 93.5% and 86.6%, respectively. In the whole enrolled population, which included patients who did not undergo surgery, the corresponding rates were 85.0% and 71.5%.

Two- and four-year disease-free survival in the surgery cohort were 83.9% and 80.4%, respectively. These survival estimates were calculated for different populations and should not be interpreted as directly comparable outcome estimates for a randomized treatment effect. The exploratory immune analysis found that patients with detectable TLS were more likely to achieve pathological complete response than patients without detectable structures, with p = 0.047.

This was an association observed in the study, not evidence that TLS caused the response or that its presence can reliably select patients for treatment. The lack of a randomized comparator, the small sample and the fact that only 31 of 40 enrolled patients proceeded to radical surgery also limit interpretation. Pathological and survival results were strongest in the surgery cohort and may therefore describe a selected group rather than every patient who began treatment.

Overall, the phase II findings provide evidence of substantial pathological response, relatively few severe treatment-related and surgical events, and encouraging survival estimates under the combined neoadjuvant and postoperative strategy, while leaving the independent contribution of toripalimab unresolved.

Subject of Research: Phase II evaluation of neoadjuvant toripalimab with chemoradiotherapy for locally advanced esophagogastric junction adenocarcinoma

Article Title: Neoadjuvant toripalimab plus concurrent chemoradiotherapy for locally advanced adenocarcinoma of the esophagogastric junction: a phase II trial

Article References: Li, Y., Du, R., Geng, J., Liu, Z., Sui, X., Zhu, X., Li, S., Zhang, Y., Ma, D., Cui, Y., Wang, H., Cai, Y., Wang, Z., Tang, L., Yuan, P., Li, S., Zhu, Y., Li, Z., Zhang, X., … Li, Z. (2026). Neoadjuvant toripalimab plus concurrent chemoradiotherapy for locally advanced adenocarcinoma of the esophagogastric junction: a phase II trial. BMC Medicine. https://doi.org/10.1186/s12916-026-05337-2

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05337-2

Keywords: toripalimab, phase II trial, esophagogastric junction cancer, neoadjuvant therapy, chemoradiotherapy, pathological response, tertiary lymphoid structures, cancer immunotherapy, single-arm study, Neoadjuvant, plus, concurrent

Yet the standard approach has a stubborn ceiling. Pathologic complete response rates with conventional neoadjuvant chemotherapy hover between roughly 6 and 16 percent, and long-term survival for patients with stage IIIA, IIIB, and IIIC disease remains only 30.5, 20.1, and 8.3 percent respectively. This gap has pushed oncologists to explore whether adding other therapeutic modalities before surgery can push more patients toward complete tumor eradication. A new systematic review and single-arm meta-analysis published in Frontiers in Oncology now offers the first quantitative synthesis of one such candidate: apatinib, an oral tyrosine kinase inhibitor of vascular endothelial growth factor receptor 2 developed in China, delivered as part of neoadjuvant regimens for resectable gastric and gastroesophageal junction adenocarcinoma.

The rationale for apatinib rests on a hallmark of cancer biology. Angiogenesis, the formation of new blood vessels, drives the growth and metastasis of aggressive tumors, with vascular endothelial growth factor and its receptors serving as key mediators. VEGFR2, one of three members of the VEGFR family, exerts a pivotal effect on VEGF-mediated tumor angiogenesis. Apatinib selectively targets this receptor and has already demonstrated efficacy in advanced disease: a phase III trial showed that apatinib improved progression-free survival from 1.8 to 2.6 months and overall survival from 4.7 to 6.5 months compared with placebo in patients with advanced gastric cancer refractory to two or more chemotherapy regimens. A phase I study also found that apatinib could enhance the efficacy of paclitaxel, oxaliplatin, and 5-FU/leucovorin in advanced gastric cancer, with objective response rates of 90 percent in the high-dose group versus 66.7 percent in the low-dose group.

To consolidate scattered preliminary clinical data, a team led by Kaiyi Xin and Zhilu Chen, with Yinghao Zhao as corresponding author, from Jilin University hospitals in Changchun, China, conducted a systematic search across PubMed, Embase, the Cochrane Library, Web of Science, and major oncology conference proceedings up to January 1, 2026. The review followed the PRISMA statement and was registered with PROSPERO under identifier CRD42023478481. The initial search identified 483 articles, from which 20 studies enrolling 1026 patients met the inclusion criteria. After excluding one potentially overlapping cohort from the same research group to avoid double-counting patients, the primary analysis rested on 19 studies with 975 patients. These comprised 11 single-arm cohort studies, 7 two-arm randomized controlled trials, and 1 three-arm randomized controlled trial, with study arms categorized as neoadjuvant apatinib combined with chemotherapy, apatinib combined with immunotherapy and chemotherapy, or apatinib combined with chemotherapy and radiotherapy. Each patient received at least two courses of neoadjuvant treatment, and most surgical resections were performed three to five weeks after the final preoperative cycle.

Because most included studies were single-arm trials reporting binary outcomes as proportions, the investigators applied a logit transformation to stabilize variance and used inverse-variance weighting to pool event rates with 95 percent confidence intervals, selecting random- or fixed-effects models based on between-study heterogeneity measured by the I-squared statistic. The headline efficacy numbers are striking. The pooled pathologic complete response rate was 21 percent (95 percent CI, 18 to 25 percent) with remarkably low heterogeneity of I-squared 19.4 percent. The pooled major pathologic response rate, a broader measure of deep tumor regression, reached 39 percent (95 percent CI, 34 to 45 percent). Complete resection with negative margins, the R0 resection rate, was achieved in 93 percent of patients (95 percent CI, 90 to 95 percent), while the pooled surgical resection rate was 92 percent. On imaging, the pooled objective response rate was 70 percent (95 percent CI, 66 to 74 percent) and the disease control rate was 95 percent (95 percent CI, 92 to 96 percent) with essentially no heterogeneity.

Safety outcomes were also quantified. The pooled incidence of grade 3 or higher treatment-related adverse events, graded according to NCI-CTCAE version 4.0, was 31 percent (95 percent CI, 21 to 43 percent), with substantial heterogeneity of I-squared 80.1 percent. Critically, no patients died as a result of these severe adverse events, and surgery was delayed in only one patient, for 93 days, because of treatment-associated pneumonia. The pooled postoperative complication rate was 27 percent (95 percent CI, 22 to 32 percent), and no patient died within 30 days of surgery from postoperative complications. Only one patient required a second operation, eventually recovering, due to bleeding from the gastrointestinal anastomosis. The most common complications, including pneumonia, hemorrhage, abdominal infection, anastomotic leakage, and pleural effusion, mostly belonged to Clavien-Dindo class II and did not dramatically affect survival prognosis.

In exploratory subgroup analyses, the pooled pathologic complete response rate was 12 percent in the apatinib plus chemotherapy group and 22 percent in the apatinib plus immunotherapy and chemotherapy group, with a descriptive estimate of 33 percent in the single apatinib plus chemoradiotherapy study. Similar directional differences appeared in major pathologic response, objective response, and R0 resection rates, which were 93.5 percent in the chemotherapy subgroup and 96.7 percent in the immunotherapy subgroup. The incidence of severe treatment-related adverse events was 23 percent in the immunotherapy subgroup versus 39 percent in the chemotherapy subgroup. The authors emphasize, however, that these are indirect, non-randomized comparisons derived from independent single-arm cohorts and are presented descriptively as hypothesis-generating, not as evidence of comparative efficacy.

The biological plausibility of combining apatinib with immune checkpoint inhibitors adds technical depth to these findings. Beyond directly suppressing tumor angiogenesis, apatinib targets components of the tumor microenvironment outside tumor cells and promotes the infiltration and activation of CD8-positive T lymphocytes. Immune checkpoint inhibitors targeting PD-1 have demonstrated antitumor activity in gastric and gastroesophageal junction adenocarcinoma, and apatinib-based regimens combined with PD-1 inhibition have shown activity in advanced disease and other solid tumors. In a three-arm randomized controlled study by Xiong and Li, the apatinib plus immunotherapy and chemotherapy group showed improved one-, two-, and three-year disease-free survival and overall survival compared with the other groups, though it caused a higher incidence of peripheral neuropathy, while the apatinib plus chemotherapy group increased the incidence of hypertension. Possible explanations include PD-1 inhibitor-induced autoreactive antibodies, cytotoxic T cells, and inflammatory cytokines affecting neural integrity, as well as apatinib-induced endothelin-1 activation and oxidative stress.

The authors also confront important limitations. The evidence base is predominantly non-comparative and heterogeneous, drawn mostly from single-arm phase II trials and retrospective cohort studies, some of which are conference abstracts without full-text peer review. Most trials have not yet reached predefined long-term endpoints, so the impact on overall and disease-free survival remains unconfirmed. Substantial clinical heterogeneity existed across studies in apatinib dose, combination regimens, number of neoadjuvant cycles, tumor stage, histology, and patient selection. Notably, the included studies generally did not stratify outcomes by histologic type or differentiation grade, and whether apatinib can overcome the relative resistance of poorly differentiated diffuse-type tumors to conventional neoadjuvant chemotherapy remains unclear. Age-related tolerability also warrants attention: a phase II study in elderly patients reported high incidences of fatigue and hypertension, and the NMPA-approved drug label in China advises caution in patients over 70 years, with individualized management including regular blood-pressure monitoring, routine urinalysis for proteinuria, and possible dose reduction to 250 mg/day for patients with poor tolerability.

Sensitivity analyses, including leave-one-out calculations and re-inclusion of the excluded overlapping cohort, confirmed that the pooled estimates remained essentially unchanged, and Egger’s regression test revealed no statistically significant publication bias for any outcome. The overall picture is one of feasibility rather than proof: apatinib-containing neoadjuvant therapy appears deliverable with an acceptable safety profile in selected patients, achieving pathologic response rates that may inform future hypothesis-driven research. The authors conclude that these descriptive estimates provide a basis for the design of future large multicenter randomized controlled trials to validate the role of apatinib in the neoadjuvant setting, while cautioning that no conclusion regarding comparative efficacy or safety superiority over conventional therapy can yet be drawn. For a cancer whose standard of care has plateaued, that call for rigorous randomized evidence may prove the most consequential finding of all.

Subject of Research: Apatinib-containing neoadjuvant therapy for resectable gastric and gastroesophageal junction adenocarcinoma

Article Title: Efficacy and safety of apatinib-containing neoadjuvant regimens for resectable gastric and gastroesophageal junction adenocarcinoma: a systematic review and single-arm meta-analysis

Article References: Xin, K., Chen, Z., Yang, L., Qin, L., Chen, S., Xiao, T., & Zhao, Y. (2026). Efficacy and safety of apatinib-containing neoadjuvant regimens for resectable gastric and gastroesophageal junction adenocarcinoma: a systematic review and single-arm meta-analysis. Frontiers in Oncology, 16, Article 1910490. https://doi.org/10.3389/fonc.2026.1910490

Image Credits: AI Generated

DOI: 10.3389/fonc.2026.1910490

Keywords: apatinib, gastric cancer, gastroesophageal junction adenocarcinoma, neoadjuvant therapy, pathologic complete response, meta-analysis, VEGFR2 inhibitor, immunotherapy, R0 resection, surgical oncology, Frontiers in Oncology, Efficacy

The study, published in Annals of Gastroenterological Surgery, drew on the nationwide Ta-Ta-Mi registry, which captured 706 patients who underwent TaTME at 26 institutions affiliated with the Japan Society of Laparoscopic Colorectal Surgery. From this pool, 679 patients with primary rectal cancer were eligible for the current analysis, all operated on between January 2012 and December 2019. The researchers focused on tumors whose distal edge lay within 12 centimeters of the anal verge—the mid and low rectum, where surgical access is most difficult and where total mesorectal excision, the meticulous removal of the fatty envelope surrounding the rectum, is critical for local tumor control.

TaTME was developed specifically to overcome the limitations of conventional laparoscopic pelvic dissection, which becomes notoriously awkward in male patients, obese patients, and those with narrow pelvises or low-lying tumors. By entering through the anus and working upward, surgeons gain improved visualization and instrument maneuverability in the deepest part of the pelvis. A network meta-analysis of 29 randomized controlled trials involving 6,237 patients previously found broadly comparable perioperative and oncological outcomes across open, laparoscopic, robotic, and transanal approaches, though laparoscopic surgery carried a higher risk of circumferential resection-margin involvement than the transanal route. What remained unclear, however, was how neoadjuvant therapy—now increasingly given as total neoadjuvant therapy, or TNT, to boost pathological complete response rates—changes the picture for TaTME specifically.

To answer that question, the research team employed propensity score matching, a statistical technique designed to mimic randomization in observational data. Patients who received neoadjuvant therapy were matched one-to-one with those who went straight to surgery, balancing the groups on age, sex, body mass index, American Society of Anesthesiologists physical status, the tumor markers CEA and CA19-9, previous abdominal surgery, and clinical tumor stage. After matching, 294 patients—147 in each arm—formed the comparison cohort, with standardized mean differences confirming good balance across nearly all prespecified covariates. The primary endpoint was severe postoperative complications, defined as Clavien–Dindo grade III or higher, a scale that grades complications from mild deviations to life-threatening events requiring invasive intervention.

The headline result was striking in its clarity on one front and its ambiguity on another. Severe complications occurred in 12.2 percent of the non-neoadjuvant group and 10.9 percent of the neoadjuvant group—an adjusted odds ratio of 0.90 with a confidence interval spanning 0.44 to 1.85, far from statistical significance. Sensitivity analyses using conventional multivariable logistic regression, Firth logistic regression, and marginal standardization on the full cohort of 676 patients likewise found no significant association. Yet the researchers were careful to note that the confidence intervals remained wide, meaning clinically relevant benefit or harm could not be excluded. The overall severe complication rate of 11.6 percent in the matched cohort sits comfortably within the 10 to 15 percent range reported in previous multicenter trials, lending external consistency to the findings.

Operative time told a different story. Patients who had received neoadjuvant therapy spent a median of 510 minutes in the operating room, compared with 382 minutes for those who had not—an adjusted mean difference of nearly 130 minutes, highly statistically significant. This association persisted even after accounting for body mass index, operative procedure, lateral pelvic lymph node dissection, and abdominal approach, with neoadjuvant therapy independently associated with a 14 percent increase in operative time. Even among patients who did not undergo lateral lymph node dissection, a procedure that itself dramatically lengthens surgery, neoadjuvant therapy remained linked to a 13.9 percent longer operation. The researchers attribute the delay to treatment-associated tissue edema, fibrosis, and less distinct anatomical planes—changes that make the deep pelvic dissection inherently more laborious, though residual confounding cannot be ruled out.

The multivariable analysis of the full cohort painted a richer portrait of what drives operative duration. Male sex, higher body mass index, previous abdominal surgery, more advanced clinical T stage, lateral lymph node dissection, abdominoperineal resection, and an open surgical approach all independently lengthened the operation, while a two-team transanal approach—simultaneous abdominal and perineal dissection—shortened it by nearly 18 percent. Notably, bilateral lateral lymph node dissection added more than 50 percent to operative time, underscoring just how much the extent of nodal surgery shapes the surgical workload, yet the absence of a significant interaction between neoadjuvant therapy and lymph node dissection extent suggests the two effects operate independently.

Pathological findings revealed the biological footprint of preoperative treatment. Lymphatic invasion was significantly less frequent in the neoadjuvant group—23.8 percent versus 43.5 percent—and venous invasion showed an even starker reduction, 38.8 percent versus 64.6 percent, both with p values below 0.001. These changes reflect the known treatment-related pathological response to neoadjuvant therapy, in which chemotherapy and radiation systematically eradicate tumor cells that have invaded lymphatic and blood vessels. Tumor budding grades also differed, with no BD3 cases in the neoadjuvant group compared with 10 in the non-neoadjuvant arm. No significant differences emerged in p/ypT category, p/ypN status, perineural invasion, or tumor differentiation.

Perhaps most clinically reassuring were the resection-margin results. Distal margin positivity was vanishingly rare in both groups—0.7 percent each—and circumferential resection-margin involvement, defined as a tumor-to-margin distance of 1 millimeter or less, occurred in 7.8 percent of evaluable non-neoadjuvant patients and 5.7 percent of neoadjuvant patients, a non-significant difference. This matters because fibrosis and tissue edema after neoadjuvant therapy could theoretically obscure the dissection plane and increase the risk of an inadequate margin. In this multicenter cohort treated at experienced centers, TaTME after neoadjuvant therapy showed no such penalty.

An unexpected observation was that anastomotic leakage—the leakage of bowel contents from a surgical connection—occurred in only 3.4 percent of neoadjuvant patients undergoing low anterior or intersphincteric resection, compared with 11.7 percent of non-neoadjuvant patients. The authors caution that this finding is exploratory: tumor regression after neoadjuvant therapy may facilitate distal rectal transection and reconstruction, but differences in patient selection, anastomotic technique, and perfusion assessment that were not captured in the database could also explain the gap, and the absolute number of leakage events was small. The regimen-specific analysis, comparing neoadjuvant chemotherapy, chemoradiotherapy, total neoadjuvant therapy, and radiotherapy alone, found no statistically significant differences in severe complications or operative time after adjustment, though the descriptive data suggested radiotherapy alone produced the longest operations.

The study carries important limitations that temper its conclusions. As a retrospective observational analysis, treatment assignment was not randomized, and the neoadjuvant group included only patients who ultimately proceeded to TaTME—those who developed disease progression or severe toxicity before surgery were not captured, meaning the findings apply to selected surgical candidates rather than the complete treatment strategy. Neoadjuvant regimen selection, lymph node dissection criteria, and sphincter-preserving indications were not standardized across institutions, the interval between treatment completion and surgery was unavailable, and pathological assessment was performed locally without centralized review. All participating centers were experienced Japanese institutions, which may limit generalizability to other healthcare systems.

Nevertheless, the nationwide multicenter design and the depth of clinical, operative, and pathological data represent substantial strengths, and the consistency of the severe-complication findings across multiple statistical approaches strengthens confidence in the central message. For surgeons and patients weighing the timing of therapy before transanal rectal cancer surgery, the evidence suggests that preoperative neoadjuvant treatment, while adding meaningful time under anesthesia, does not appear to compromise the safety profile or the quality of the oncological resection in experienced hands. As total neoadjuvant therapy continues to expand globally in the pursuit of higher pathological complete response rates and improved systemic disease control, these findings provide a practical foundation for surgical planning in the era of increasingly intensive preoperative treatment.

Subject of Research: Short-term surgical outcomes of transanal total mesorectal excision for rectal cancer with and without preoperative neoadjuvant therapy

Article Title: Short‐Term Outcomes of Transanal Total Mesorectal Excision for Rectal Cancer: A Multicenter Propensity Score‐Matched Analysis Comparing Patients With and Without Preoperative Neoadjuvant Therapy

Article References: Shima, T., Hamamoto, H., Matsuda, T., Takemasa, I., Tanaka, K., Mukai, T., Watanabe, J., Lee, S.-W., Yamamoto, S., Naitoh, T., & Ta‐Ta‐Mi Study Group Collaborators from the Japan Society of Laparoscopic Colorectal Surgery (2026). Short‐Term Outcomes of Transanal Total Mesorectal Excision for Rectal Cancer: A Multicenter Propensity Score‐Matched Analysis Comparing Patients With and Without Preoperative Neoadjuvant Therapy. Annals of Gastroenterological Surgery, Article ags3.70298. https://doi.org/10.1002/ags3.70298

Image Credits: AI Generated

DOI: 10.1002/ags3.70298

Keywords: rectal cancer, transanal total mesorectal excision, neoadjuvant therapy, total mesorectal excision, propensity score matching, surgical outcomes, anastomotic leakage, Clavien-Dindo complications, chemoradiotherapy, total neoadjuvant therapy, operative time, circumferential resection margin

HER2-positive tumors account for 20 to 25 percent of all breast cancer cases and are among the most aggressive forms of the disease, associated with poorer outcomes when not effectively treated. Neoadjuvant therapy is frequently administered before surgery with the goal of shrinking the tumor, sometimes to the point where surgeons find no residual invasive cancer at all—a result known as pathologic complete response, which carries important prognostic weight. The problem, according to the researchers behind the new study, is that the tools currently used to monitor treatment response each fall short in some way.

“Ultrasound offers cost-effective assessment of lymph nodes but is operator-dependent; mammography detects calcifications well but poorly measures tumor size; MRI, the standard neoadjuvant therapy monitoring modality, faces challenges in differentiating between fibrosis and tumors; and clinical examination, although rapid, lacks accuracy,” said Yunjiang Liu, MD, PhD, professor at the Fourth Hospital of Hebei Medical University and Hebei Provincial Tumor Hospital in Shijiazhuang, China. “This trade-off between accuracy and timeliness drives the need for novel approaches.”

That is where molecular imaging enters the picture. 68Ga-HER2 PET/CT uses a radioactive tracer engineered to bind specifically to HER2 receptors on the surface of cancer cells, allowing physicians to visualize HER2 expression throughout the entire body noninvasively. Unlike anatomical imaging, which measures the size of a tumor, molecular imaging measures biological activity—how much of the HER2 target is actually present and accessible. When treatment begins to work, HER2 expression and tumor metabolism can change before the tumor itself measurably shrinks, potentially offering an earlier window into therapeutic success or failure.

To test this concept rigorously, the research team conducted a prospective study of 44 patients with biopsy-confirmed HER2-positive breast cancer. Every patient received at least four cycles of neoadjuvant therapy, though the specific regimens varied from patient to patient, followed by surgery. The researchers performed 68Ga-HER2 PET/CT scans at three time points: at baseline before treatment began, after the second therapy cycle, and again before surgery. They then measured how tracer uptake in the primary breast tumors and in lymph nodes changed across these imaging sessions and analyzed whether those changes correlated with pathologic complete response when the surgical specimens were examined under the microscope.

The results were striking in one respect and sobering in another. The baseline scan, performed before any treatment, did not distinguish between patients who would go on to achieve a pathologic complete response and those who would not. But after only two cycles of therapy, a clear signal emerged. Patients whose primary tumor tracer uptake dropped by at least 61 percent after the second treatment cycle had an 85 percent rate of pathologic complete response at surgery. Among patients whose uptake fell by less than that threshold, the complete response rate was just 38 percent. In other words, a single early scan nearly doubled the predictive separation between responders and non-responders.

The imaging findings from a representative case illustrate the principle vividly. At baseline, one patient’s scan showed a soft-tissue lesion measuring approximately 2.3 by 2.2 centimeters in the upper outer quadrant of the left breast, with marked radiotracer uptake and a maximum standardized uptake value of 10.1—a measure of how intensely the tracer accumulated in the tumor. At the end of the second treatment cycle, the residual lesion measured approximately 2.2 by 1.0 centimeters, but crucially, no abnormal radiotracer accumulation remained. By the preoperative scan, the lesion had shrunk further to approximately 2.0 by 0.6 centimeters, again with no abnormal tracer uptake detected. The biological signal had vanished well before the anatomical lesion did.

“These findings provide preliminary prospective evidence that changes in HER2-targeted tracer uptake can predict treatment response at a very early stage,” said Xinming Zhao, MD, PhD, director of the Department of Nuclear Medicine at the Fourth Hospital of Hebei Medical University and Hebei Provincial Tumor Hospital. “For patients, the potential importance is that treatment response may be assessed much earlier. If future studies confirm our findings, imaging could provide earlier information to support discussions about alternative treatment strategies or surgical timing.”

The clinical implications could be substantial. If an early 68Ga-HER2 PET/CT scan shows that a patient’s tumor is not responding—tracer uptake remaining stubbornly high after two cycles—physicians could consider switching to an alternative treatment strategy sooner, or move directly to surgery rather than continuing a regimen that appears ineffective. Conversely, patients showing a dramatic drop in uptake could proceed with confidence that their current therapy is working. This kind of early, biologically grounded feedback addresses precisely the trade-off that Liu described: the existing modalities are either accurate but slow, or fast but imprecise.

The authors are careful to note the limitations of their work. The study was small, conducted at a single center, and included variation in neoadjuvant treatment regimens among patients. Moreover, the study was designed to evaluate diagnostic performance only—it did not test whether actually changing treatment based on PET/CT findings improves patient outcomes, which would require a different kind of trial. Those caveats mean the 61 percent uptake-reduction threshold and the response rates reported here should be considered preliminary rather than definitive. Still, as a prospective study with serial imaging and surgical confirmation, it provides some of the strongest evidence to date that HER2-targeted molecular imaging can serve as an early biomarker of treatment response. Larger, multicenter trials will be needed to validate the findings and determine whether early 68Ga-HER2 PET/CT can genuinely change the course of care for patients with this aggressive form of breast cancer.

Subject of Research: Using 68Ga-HER2 PET/CT molecular imaging to predict pathologic complete response in HER2-positive breast cancer during neoadjuvant therapy

Article Title: Molecular imaging demonstrates promise for predicting complete pathologic response in HER2-positive breast cancer

Article References: Molecular imaging demonstrates promise for predicting complete pathologic response in HER2-positive breast cancer. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: HER2-positive breast cancer, 68Ga-HER2 PET/CT, molecular imaging, neoadjuvant therapy, pathologic complete response, PET imaging, radiotracer, breast cancer treatment, nuclear medicine, treatment response prediction, oncology, The Journal of Nuclear Medicine

Radiomics is the high-throughput extraction of quantitative features from medical images, converting scans into hundreds of numerical descriptors that capture tumor characteristics invisible to the naked eye. When applied to 18F-fluorodeoxyglucose positron emission tomography combined with computed tomography, radiomic features include first-order statistics, texture matrices such as the gray-level co-occurrence, run-length, size-zone, and neighborhood gray-tone difference matrices, and morphological descriptors. These measurements may encode intratumoral heterogeneity that conventional semiquantitative parameters like maximum standardized uptake value, metabolic tumor volume, and total lesion glycolysis cannot capture. Because PET/CT is already integral to esophageal cancer staging, radiomics offers a non-invasive, automated route to tumor phenotyping that could transform treatment planning.

The critical insight motivating the new analysis is that radiomics can be applied at different points along the treatment trajectory, and the imaging time point is itself a biologically and methodologically important determinant of predictive performance. Baseline, pre-treatment radiomics characterizes the intrinsic biology of the untreated tumor. Restaging, post-treatment radiomics characterizes the residual metabolic phenotype but is confounded by radiation-induced inflammation that can mimic residual tumor. Delta radiomics quantifies the longitudinal change between paired scans, potentially isolating the treatment effect itself. Existing single-center studies vary widely in sample size, imaging time point, radiomic methodology, and validation strategy, and no quantitative synthesis had compared these three time points directly until now.

A team led by Jie Li and Subapriya Suppiah of Universiti Putra Malaysia, with collaborators from Beacon Hospital, Zhongshan Hospital Fudan University, Austin Health in Melbourne, and King Hussein Cancer Center in Amman, conducted the review in accordance with PRISMA-DTA guidelines and prospectively registered it with PROSPERO. The researchers searched PubMed/MEDLINE, the Cochrane Library, Scopus, and EBSCO from inception to March 2026 without language restriction. Eligible studies enrolled adults with histologically confirmed esophageal squamous cell carcinoma or esophageal adenocarcinoma scheduled for neoadjuvant chemoradiotherapy or chemoimmunotherapy followed by surgical resection, applied PET/CT radiomics at one or more time points, and used surgical histopathological examination as the reference standard for strictly defined pathological complete response.

Thirteen publications were ultimately included, generating 14 study-time-point observations, with one cohort contributing to two temporal pools. The included studies, published between 2013 and 2026, originated from the United States, the Netherlands, Taiwan, China, and Japan, with sample sizes ranging from 20 to 189 patients and reported pathological complete response rates from 20.0 to 59.2 percent. Modeling approaches ranged from univariate ROC analysis and multivariable logistic regression to support vector machines and deep-learning combined models incorporating habitat-aware radiomics and multiple-instance learning. Only two studies employed external validation, while the remainder relied on internal validation strategies such as bootstrap resampling, split-half designs, leave-one-out cross-validation, or five-fold cross-validation.

The headline quantitative result concerns the baseline pool. Eight pre-treatment studies totaling 564 patients yielded a pooled area under the receiver operating characteristic curve of 0.845, with a 95 percent confidence interval of 0.768 to 0.899 and moderate heterogeneity quantified at an I-squared of 34.6 percent. Individual study AUCs ranged from 0.74 to 0.95, with the highest estimate coming from a Japanese neural-network classifier applied to squamous cell carcinoma patients. By contrast, the four post-treatment studies totaling 387 patients produced a pooled AUC of just 0.677, with an extraordinarily wide confidence interval spanning 0.328 to 0.900 and considerable heterogeneity at I-squared of 82.8 percent. Individual post-treatment estimates ranged from 0.42 to 0.85, with the only externally validated entry achieving just 0.64, a well-recognized shrinkage of apparent performance when models are tested on new data.

The delta pool, comprising only two studies with 65 patients, was summarized descriptively rather than pooled because the minimum threshold of four studies per pool was not met. The individual delta AUCs were 0.89 and 0.79, both derived from small single-center cohorts using texture and histogram-distance features. The authors note that an illustrative random-effects interval for the delta pool would span 0.04 to 1.00, rendering any quantitative pooled estimate uninformative. They characterize the delta findings as hypothesis-generating only, though conceptually attractive because delta radiomics isolates treatment-induced change in tumor phenotype and may be less sensitive to inter-scanner differences in absolute feature values.

An exploratory subgroup analysis restricted to baseline estimates compared radiomics-only models with models combining radiomics and clinical variables. The radiomics-only subgroup of six studies yielded a pooled AUC of 0.831, while the radiomics-plus-clinical subgroup of four studies achieved 0.814, a negligible between-subgroup difference of 0.017 that was not statistically significant. However, within-cohort paired analyses told a different story: in individual studies, adding clinical variables increased the AUC from 0.71 to 0.852 and from 0.772 to 0.824. The authors caution that with so few studies, these exploratory results should be interpreted with care, and publication bias could not be reliably assessed because fewer than 10 studies were available in each pool, precluding funnel-plot analysis in accordance with the Cochrane Handbook.

Methodological quality assessment revealed both reassurance and concern. Under the QUADAS-2 tool, eight study entries achieved an overall low risk of bias and five were rated unclear, with no entry rated high risk. The most frequent source of uncertainty was the index-test domain, reflecting infrequent reporting of blinding of radiomic model development and segmentation to the pathological outcome, together with thresholds derived from the same dataset used for model training. The Radiomics Quality Score, which assesses radiomics-specific dimensions across 16 components with a maximum of 36 points, was moderate overall with a median of 44.4 percent and individual scores from 19 to 67 percent. Recurrent deficiencies included phantom and inter-observer reproducibility testing, compliance with the Image Biomarker Standardization Initiative, prospective registration, and open-science practices, shortcomings that could prevent reliable deployment in multicenter practice.

The findings carry direct clinical implications for organ-preservation strategies. The preSANO and SANO trials demonstrated that selective non-operative management can be oncologically safe in patients with clinical complete response after neoadjuvant chemoradiotherapy, creating an urgent need for reliable pre-treatment biomarkers to triage candidates. A baseline radiomic model with consistent accuracy near 0.85 could complement endoscopic and clinical assessment to stratify patients before therapy, intensifying treatment in predicted non-responders and prioritizing predicted responders for closer surveillance, potentially in combination with circulating tumor DNA monitoring. The emergence of chemoimmunotherapy adds another dimension: three baseline-pool studies in the immune-checkpoint-inhibitor setting achieved high accuracy with AUCs of 0.852, 0.87, and 0.824, and the highest pathological complete response rates in the dataset, consistent with the additive antitumor effect of PD-1 blockade. The authors suggest that features of the tumor immune microenvironment driving checkpoint response, such as CD8-positive T-cell infiltration and PD-L1 expression, may be partially encoded in pre-treatment FDG PET radiomic features through their association with metabolic activity and intratumoral heterogeneity.

Important caveats temper these encouraging signals. All included studies were retrospective and predominantly single-center, with only two employing external validation. The baseline and post-treatment estimates were derived from different study sets and should not be interpreted as a direct head-to-head comparison. Acquisition protocols differed across studies in injected dose, uptake time, and acquisition duration, potentially affecting feature comparability, and each pooled AUC summarizes methodologically diverse models rather than a single standardized signature. The authors conclude that multicenter prospective studies employing standardized, IBSI-compliant radiomic workflows, which compare baseline, delta, and post-treatment imaging directly within the same cohorts, represent the critical next step toward clinical translation, alongside dedicated model development for chemoimmunotherapy regimens, integration with liquid-biopsy and tissue biomarkers, and health-economic evaluation of radiomic-guided treatment stratification.

Subject of Research: Diagnostic accuracy of 18F-FDG PET/CT radiomics across baseline, restaging, and delta imaging timepoints for predicting pathological complete response to neoadjuvant therapy in locally advanced esophageal cancer

Article Title: 18F-FDG PET/CT radiomics across treatment timepoints for predicting response to neoadjuvant therapy in locally advanced esophageal cancer: a systematic review and meta-analysis

Article References: Li, J., Suppiah, S., Kasilingam, L., Hui, S. T., Shi, H., Lee, S. T., Sharin, M. F. B. M., & Al-Ibraheem, A. (2026). 18F-FDG PET/CT radiomics across treatment timepoints for predicting response to neoadjuvant therapy in locally advanced esophageal cancer: a systematic review and meta-analysis. Frontiers in Medical Engineering, 4, Article 1979078. https://doi.org/10.3389/fmede.2026.1979078

Image Credits: AI Generated

DOI: 10.3389/fmede.2026.1979078

Keywords: 18F-FDG PET/CT, radiomics, esophageal cancer, pathological complete response, neoadjuvant therapy, meta-analysis, systematic review, delta radiomics, immune-checkpoint inhibitors, machine learning, treatment response prediction, medical imaging

The research, conducted by a team at the University of Health Sciences Prof. Dr. Cemil Tascioglu City Hospital and published in BMC Cancer, examined 138 premenopausal women with non-metastatic, hormone receptor-positive, HER2-positive breast cancer. All of the patients had undergone neoadjuvant therapy — treatment given before surgery — built around anti-HER2 agents, followed by surgical removal of their tumors between 2015 and 2024. The investigators then split the cohort according to whether the women received ovarian function suppression, or OFS, during the adjuvant period, the phase of care that follows surgery and is designed to mop up any remaining cancer cells.

The biology behind the question is straightforward. In premenopausal women, the ovaries are the dominant source of estrogen, the hormone that fuels the growth of hormone receptor-positive tumors. Blocking ovarian output — typically with gonadotropin-releasing hormone agonists that effectively switch the ovaries off — dramatically lowers circulating estrogen. Combined with drugs such as tamoxifen, which blocks estrogen receptors on cancer cells, or aromatase inhibitors, which block estrogen synthesis in peripheral tissues, ovarian suppression is intended to starve any residual disease of its principal growth signal. In HER2-negative disease, landmark trials have shown clear benefits from this strategy. But HER2-positive tumors behave differently, and the arrival of potent anti-HER2 therapies such as trastuzumab has transformed outcomes, raising the possibility that the incremental value of ovarian suppression has shrunk.

That possibility is precisely what the Turkish team set out to test. Of the 138 patients, 60 — or 43.5 percent — received ovarian function suppression as part of their adjuvant endocrine treatment. The remaining 78 did not. The researchers tracked three primary endpoints: disease-free survival, which measures the time before cancer returns anywhere in the body; distant disease-free survival, which focuses specifically on spread to distant organs; and overall survival. They also assessed pathological complete response, abbreviated pCR, as a secondary endpoint. A pCR means that when pathologists examine the tissue removed at surgery, they find no invasive cancer in the breast and no cancer in the sampled lymph nodes — a marker widely regarded as one of the strongest predictors of long-term cure.

The headline numbers were striking in places. Overall, 42 percent of the entire cohort achieved a pathological complete response, a figure that underscores just how powerful neoadjuvant anti-HER2 therapy has become. After a median follow-up of 56 months, 20 patients — 14.5 percent — had experienced a recurrence or developed metastatic disease, and 11 patients, or 8 percent, had died. Those figures, sobering as they are, reflect the reality of a disease subtype that remains challenging even in the modern therapeutic era.

When the investigators compared the two groups, the differences were modest at best. Five-year overall survival was 93.9 percent among women who received ovarian suppression versus 85.8 percent among those who did not — a numerical gap, but one that did not reach statistical significance, with a p-value of 0.4. Disease-free survival was essentially identical between the arms: 80.7 percent with ovarian suppression and 81.3 percent without, yielding a p-value of 0.9. Distant disease-free survival followed the same pattern. In other words, on the primary measures of whether cancer came back or whether patients lived, the addition of ovarian suppression made no statistically detectable difference in this cohort.

The most intriguing signal emerged when the researchers stratified patients by their response to neoadjuvant therapy. Women who achieved a pathological complete response had excellent outcomes regardless of whether their ovaries were subsequently suppressed — their five-year overall survival was a remarkable 100 percent. This finding reinforces a growing consensus in oncology: for patients whose tumors vanish under preoperative anti-HER2 treatment, the prognosis is already so favorable that additional endocrine intensification may contribute little. The residual-disease group told a more nuanced story. Among women who still had detectable cancer at surgery, ovarian suppression was associated with numerically higher overall survival — 89.9 percent versus 77.6 percent — but this apparent advantage did not translate into a disease-free survival benefit, and the difference was not statistically significant.

To account for the possibility that the two groups differed in ways that could confound the comparison — differences in tumor size, nodal involvement, grade, or the type of endocrine therapy received — the team performed multivariate analysis, a statistical technique that adjusts for multiple variables simultaneously. The result was unambiguous: ovarian function suppression showed no independent association with disease-free survival, distant disease-free survival, or overall survival once other factors were taken into account. The authors are careful to note, however, that the study’s statistical power was limited. With 138 patients and relatively few events — 20 recurrences and 11 deaths — the analysis simply cannot exclude a modest benefit that a larger trial might detect.

Another critical caveat concerns the composition of the endocrine therapy itself. The cohort was predominantly treated with tamoxifen, the selective estrogen receptor modulator that has long been the backbone of adjuvant therapy in premenopausal women. Whether ovarian suppression paired with aromatase inhibitors — a more aggressive regimen that eliminates nearly all estrogen production — would perform differently cannot be answered by this dataset. The authors explicitly caution that, given the limited statistical power and the tamoxifen-dominated treatment landscape, their findings should be interpreted with care, and they call for prospective studies to define the role of ovarian suppression in this population with the rigor that only randomized designs can provide.

For clinicians and patients, the practical message is one of individualized decision-making rather than a blanket rule. Ovarian suppression is not a free intervention: it induces a premature, often abrupt menopause, bringing hot flashes, bone density loss, sexual dysfunction, and potential effects on fertility and quality of life. If a young woman achieves a pathological complete response after neoadjuvant anti-HER2 therapy, the data suggest her prognosis is excellent without intensification, and she and her physician may reasonably weigh the burden of suppression against its uncertain benefit. For those with residual disease, the numerically better overall survival observed with suppression — though not statistically significant — may still inform a shared decision, particularly for patients with other high-risk features. What this single-center experience ultimately demonstrates is that the era of effective anti-HER2 therapy has changed the calculus of endocrine intensification, and that the questions guiding treatment for premenopausal, hormone receptor-positive, HER2-positive breast cancer are now as much about avoiding overtreatment as about maximizing it. Larger, prospective trials will be needed to settle the matter definitively, but for now, this study offers a data-driven reminder that more intensive therapy is not always better therapy.

Subject of Research: Ovarian function suppression in adjuvant endocrine therapy for premenopausal hormone receptor-positive, HER2-positive breast cancer after neoadjuvant anti-HER2 therapy

Article Title: The effect of adding ovarian function suppression to adjuvant endocrine therapy in premenopausal patients with hormone receptor-positive, HER2-positive breast cancer who have undergone neoadjuvant therapy: a single-center experience

Article References: Aydin, O., Dogan, H. G. Y., Sonusen, S. D., Erciyestepe, M., Ozturk, A. E., Buyukkuscu, A., Isleyen, Z. S., & Erturk, K. (2026). The effect of adding ovarian function suppression to adjuvant endocrine therapy in premenopausal patients with hormone receptor-positive, HER2-positive breast cancer who have undergone neoadjuvant therapy: a single-center experience. BMC Cancer, 26(1), Article 1140. https://doi.org/10.1186/s12885-026-17017-8

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17017-8

Keywords: breast cancer, HER2-positive, hormone receptor-positive, ovarian function suppression, neoadjuvant therapy, adjuvant endocrine therapy, pathological complete response, disease-free survival, overall survival, tamoxifen, premenopausal, oncology

The research team assembled five complementary streams of data from sixty-three women with histologically confirmed invasive breast carcinoma who underwent baseline fluorine-18 fluorodeoxyglucose positron-emission tomography combined with computed tomography, the scan commonly abbreviated as 18F-FDG PET/CT. Forty-two of the patients were node-positive and twenty-one were node-negative, and forty of the sixty-three, roughly sixty-three percent, subsequently received neoadjuvant systemic therapy. From the tumour volume delineated on co-registered PET and CT images, the investigators extracted 2,167 radiomics features compliant with the Image Biomarker Standardisation Initiative, 12,288 deep-learning features obtained through transfer learning from a ResNet-50 convolutional neural network, eleven volumetric metabolic parameters, seven routine clinicopathological variables and seven blood-count-derived indices of systemic inflammation and nutritional status.

What distinguishes this study from many similar multimodal efforts is not the ambition of its feature set but the rigour of its evaluation. In machine-learning medicine, a notorious pitfall is information leakage: when preprocessing steps such as feature selection are fitted on the entire dataset before validation, the model inadvertently peeks at the very cases it is supposed to predict, inflating apparent performance. The authors designed their pipeline to prevent this. Median imputation of missing values, variance filtering, standardisation, univariate screening by area under the receiver-operating-characteristic curve, collinearity pruning and least absolute shrinkage and selection operator, or LASSO, feature selection were all fitted exclusively within the training folds of repeated stratified five-fold cross-validation, repeated ten times.

The modelling search itself was exhaustive. Five single-block models, one for each data stream, were compared against three fusion strategies: stacking, feature concatenation and a parameter-free probability mean. Four base learners were tested, and critically, the complete choice of data block, fusion strategy, learning algorithm and decision threshold was repeated inside a nested cross-validation scheme. Uncertainty was quantified by patient-level bootstrap, models were compared using paired bootstrap tests, and the authors additionally assessed calibration, decision-curve net benefit, the incremental value of imaging over clinicopathological variables, the handling of missing data and robustness to contour perturbation of the tumour outlines.

The headline result is sobering but honest. The nested cross-validated area under the curve of the entire modelling procedure was 0.711, with a ninety-five percent confidence interval of 0.609 to 0.805. The probability-mean fusion, the strategy most frequently selected in the inner loops of the nested validation, reached 0.795 on the same splits, with a confidence interval of 0.678 to 0.891. The gap between these two figures, approximately 0.08, is a transparent quantification of model-selection optimism, the amount by which the best-performing configuration overstates true performance simply because it was chosen from many candidates. At thresholds set on training folds, the fused model achieved a sensitivity of 0.80 and a specificity of 0.60, and its calibration was imperfect, with a slope of 1.53, meaning predicted probabilities did not align perfectly with observed frequencies.

Perhaps the most clinically provocative finding concerns the value of imaging itself. The clinicopathological variables alone achieved a cross-validated AUC of 0.681. Adding the PET/CT-derived information increased the AUC by 0.107, but the confidence interval for that increment ranged from minus 0.023 to 0.247, and the paired comparison yielded a p-value of 0.12. In plain terms, the study could not statistically demonstrate that the expensive imaging features added incremental predictive value over routinely collected clinical data in this cohort. Decision-curve analysis reinforced the caution: the fused model offered a net benefit over a treat-all strategy only at threshold probabilities of about 0.4 or higher, a relatively narrow window of clinical usefulness.

The single-block results add further texture. Deep-learning features extracted by the ResNet-50 network were the strongest individual predictor, reaching an AUC of 0.750, while the haematological inflammation-nutrition indices performed worst at 0.590, a confidence interval that included 0.5, the level of a coin flip. This hierarchy is biologically plausible, since deep convolutional networks can capture texture and morphological patterns in tumour glucose metabolism that correlate with aggressive biology and nodal spread, whereas systemic inflammation markers reflect host factors that are only loosely coupled to lymphatic dissemination.

The authors also stress-tested their features in ways that many radiomics studies neglect. Under a stringent one-voxel perturbation of the tumour contours, only twenty-three percent of the radiomics features showed an intraclass correlation coefficient of at least 0.75, a standard benchmark for reproducibility. Importantly, restricting the radiomics block to these robust features did not reduce model performance, suggesting that the pipeline can be pruned to stable features without loss. Discrimination was notably poor in the subgroup of patients who received neoadjuvant systemic therapy, where the nested AUC fell to 0.561, an expected consequence of chemotherapy altering both tumour metabolism and imaging appearance before surgery.

Why does a modest result merit attention? Because the study is a methodological exemplar in a field awash with overclaimed performance. Many published radiomics models report dazzling AUCs that evaporate under external validation, largely because of leakage, small samples and optimistic model selection. By embedding every preprocessing and selection decision inside nested cross-validation, quantifying selection optimism explicitly, reporting calibration alongside discrimination and publishing decision-curve analyses, the Chinese team has provided a template for how multimodal prediction models should be evaluated. The result, a moderate AUC of roughly 0.71 to 0.80 with imperfect calibration, is likely a far more accurate reflection of what such models can currently achieve than the near-perfect figures that populate the literature.

The authors are explicit that this work should be regarded as a proof-of-concept framework rather than a clinical tool. The cohort was small, retrospective and drawn from a single centre, the First Affiliated Hospital of Henan University of Science and Technology in Luoyang, and the study was approved by that institution’s medical ethics committee with the requirement for written informed consent waived owing to the retrospective design. Before any clinical application, prospective multicentre external validation is required, including direct comparison with routine visual PET/CT assessment of the axilla. If such validation confirms even the moderate performance observed here, the payoff could be substantial: a non-invasive preoperative estimate of nodal status could help select patients for sentinel biopsy versus more extensive axillary surgery, spare node-negative women unnecessary procedures and inform neoadjuvant treatment decisions. For now, the study’s greatest contribution may be cultural, demonstrating that in the race to bring artificial intelligence into cancer staging, honesty about uncertainty is not a weakness but the very foundation of trustworthy medicine.

Subject of Research: Preoperative prediction of axillary lymph-node metastasis in invasive breast cancer using a leakage-controlled multimodal machine-learning model based on 18F-FDG PET/CT radiomics, deep-learning features, metabolic parameters and haematological inflammation-nutrition indices

Article Title: A leakage-controlled multimodal model integrating 18F-FDG PET/CT radiomics, deep-learning features, metabolic parameters and haematological inflammation–nutrition indices for preoperative prediction of axillary lymph-node metastasis in invasive breast cancer

Article References: Nie, H., Tan, H., Cao, Z., & Cao, X. (2026). A leakage-controlled multimodal model integrating 18F-FDG PET/CT radiomics, deep-learning features, metabolic parameters and haematological inflammation–nutrition indices for preoperative prediction of axillary lymph-node metastasis in invasive breast cancer. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02948-2

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02948-2

Keywords: breast cancer, axillary lymph-node metastasis, 18F-FDG PET/CT, radiomics, deep learning, machine learning, nested cross-validation, data leakage, biomarkers, cancer staging, predictive medicine, neoadjuvant therapy

The concept of textbook outcome is deceptively simple. It is a composite quality benchmark that is satisfied only when every element of a surgical episode goes ideally: the operation achieves an R0 resection, meaning the tumor and its margins are completely removed with no residual cancer visible to the pathologist; an adequate number of lymph nodes is dissected and examined so that staging is reliable; and the perioperative period passes without significant complications, readmission, or prolonged hospitalization. If any single component fails—a positive margin, an insufficient nodal sample, an anastomotic leak, a prolonged air leak requiring reintervention—the textbook outcome is not achieved. The measure originated in surgical oncology as a way to capture quality of care in a single, patient-centered metric rather than a scatter of separate indicators, and it has since been validated in esophageal, pancreatic, colorectal, and gastric cancer surgery. Its application to non-small cell lung cancer treated with neoadjuvant therapy, however, had remained underexplored until now.

The study team, led by researchers at Fujian Medical University Union Hospital together with collaborators at six tertiary hospitals across Fujian Province, China, retrospectively analyzed 347 patients with non-small cell lung cancer who underwent curative-intent resection between 2015 and 2022 after receiving neoadjuvant therapy. Neoadjuvant treatment—chemotherapy, immunotherapy, or chemoradiotherapy delivered before surgery—has become a cornerstone for locally advanced disease, increasingly incorporating immune checkpoint inhibitors that target the PD-1 and PD-L1 pathways. But the preoperative drugs add complexity: they can alter tissue planes, inflame the operative field, and influence both pathological response and surgical difficulty. The researchers asked whether the composite textbook outcome measure could cut through this complexity and deliver prognostic information that conventional measures miss.

The headline finding was that 278 of the 347 patients, or 80.1 percent, achieved a textbook outcome. That figure is itself informative: even in experienced tertiary centers, roughly one in five patients treated with neoadjuvant therapy before lung resection experienced at least one deviation from the ideal surgical pathway. Using logistic regression, the team identified two independent predictors of failure. The first was a Charlson comorbidity index of 3 or higher, a weighted score that quantifies a patient’s burden of pre-existing conditions such as severe cardiac disease, chronic pulmonary disease, diabetes with end-organ damage, or renal impairment. The second was ypN-positive disease—evidence that cancer had persisted in the regional lymph nodes despite the preoperative treatment. Both factors make intuitive sense. Patients with multiple comorbidities have less physiological reserve to weather surgical stress, while residual nodal disease often reflects more aggressive biology and more extensive, technically demanding dissections.

The survival analysis yielded the study’s most consequential result. After multivariable adjustment for competing risk factors, failure to achieve a textbook outcome was independently associated with shorter overall survival, with a hazard ratio of 1.61 and a 95 percent confidence interval of 1.02 to 2.52, reaching statistical significance at P equal to 0.039. In practical terms, patients whose surgical course deviated from the textbook ideal faced roughly a 60 percent higher hazard of death during follow-up compared with those who achieved it. The association with disease-free survival, by contrast, did not reach statistical significance, with a hazard ratio of 1.33 and a confidence interval spanning 0.88 to 2.01 at P equal to 0.180. The authors interpret this asymmetry carefully: the composite metric appears to capture not only oncological completeness but also the physiological toll of a complicated recovery, which may weigh more heavily on long-term overall survival than on the timing of cancer recurrence alone.

Perhaps the most clinically provocative finding concerned adjuvant therapy delivery. Postoperative treatment—additional chemotherapy, radiation, or immunotherapy given after recovery from surgery—is a critical lever for improving outcomes in patients at high risk of recurrence. Yet the study found that patients who achieved a textbook outcome were far more likely to receive adjuvant therapy than those who did not: 72.3 percent versus 46.4 percent, a difference that was highly significant. Even after statistical adjustment, failure to achieve a textbook outcome remained strongly associated with a lower likelihood of receiving adjuvant treatment, with an odds ratio of 0.28. This creates what the researchers describe as a concerning cascade: the very patients whose complicated recoveries signal higher risk are also the patients most likely to fall through the cracks of postoperative treatment, whether because of prolonged convalescence, clinician hesitancy, patient frailty, or logistical barriers that compound one another.

The study also tackled a question at the frontier of thoracic oncology: how the textbook outcome measure relates to major pathological response, or MPR. MPR is defined as the presence of 10 percent or fewer viable tumor cells in the resected specimen and has emerged as a leading surrogate endpoint in trials of neoadjuvant immunotherapy, because it reflects how completely the preoperative drugs eradicated the cancer. But MPR says nothing about the surgery itself or the recovery. The researchers therefore constructed a combined stratification that crossed MPR status with textbook outcome status. The analysis revealed distinct prognostic groups—patients who achieved both fared best, while discordant groups occupied intermediate positions—demonstrating that pathological response and perioperative recovery carry complementary, non-overlapping information.

The statistical evidence for that complementarity was formalized through model-fit testing. Adding textbook outcome to a prognostic model built on MPR alone significantly improved the fit for both overall survival and disease-free survival, with P values of 0.004 and 0.029 respectively, as assessed by likelihood-based criteria including the Akaike information criterion. In plain language, a model that knows only how the tumor responded to drugs is measurably less informative than one that also knows whether the operation and recovery proceeded without incident. This argues for a two-axis approach to postoperative risk assessment: one axis describing the biology of the residual disease, the other describing the quality and consequences of the surgical episode itself.

The implications for clinical practice are tangible. First, textbook outcome could serve as a practical quality-improvement target for thoracic surgery programs, since its component criteria—R0 resection, adequate nodal dissection, uncomplicated recovery—are auditable in routine records. Second, the strong link between failing to achieve the composite endpoint and reduced adjuvant therapy delivery suggests a concrete intervention point: patients who experience perioperative complications or incomplete resections should be flagged for dedicated follow-up to ensure that adjuvant treatment decisions are actively revisited once they recover, rather than quietly abandoned. Third, the combined MPR–textbook outcome framework offers a more nuanced postoperative risk stratification tool that could guide trial design and, eventually, individualized surveillance intensity.

The authors and commentators alike caution that the retrospective design imposes limits. The cohort was drawn from six tertiary hospitals in a single Chinese province, and patterns of perioperative care, adjuvant treatment selection, and follow-up may differ elsewhere. Retrospective analyses are also vulnerable to residual confounding: unmeasured factors—tumor molecular characteristics, specific neoadjuvant regimens, surgeon volume, patient preferences—could influence both the likelihood of a textbook outcome and survival. The authors note that the study was approved by the Institutional Ethics Review Board of Fujian Medical University Union Hospital and conducted under the Declaration of Helsinki, with informed consent waived for the anonymized retrospective data. Prospective validation in diverse populations, and ideally in the context of ongoing immunotherapy-era trials, will be needed before textbook outcome becomes a standard element of postoperative decision-making in lung cancer. Still, the study makes a compelling case that the road to better lung cancer outcomes runs not only through shrinking tumors before surgery, but through ensuring that the operation and its aftermath unfold exactly as the textbook says they should.

Subject of Research: Prognostic value of textbook outcome in resected non-small cell lung cancer after neoadjuvant therapy

Article Title: Prognostic value of textbook outcome in resected non-small cell lung cancer after neoadjuvant therapy: a multicenter retrospective study

Article References: Huang, J., Lin, T.-Z., Wang, Z.-J., He, L.-P., Tu, J.-H., Lian, L.-Y., Jin, Z.-W., Wang, H.-J., Shen, G.-Y., Xu, S.-J., & Chen, S.-C. (2026). Prognostic value of textbook outcome in resected non-small cell lung cancer after neoadjuvant therapy: a multicenter retrospective study. BMC Cancer. https://doi.org/10.1186/s12885-026-17131-7

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17131-7

Keywords: non-small cell lung cancer, textbook outcome, neoadjuvant therapy, major pathological response, adjuvant therapy, overall survival, disease-free survival, R0 resection, lymph node dissection, perioperative complications, prognosis, thoracic surgery

Neoadjuvant therapy, the treatment given before a tumor is surgically removed, has become a cornerstone of modern cancer care. In rectal, breast, esophageal, lung, and many other cancers, shrinking the tumor with chemotherapy, radiotherapy, or immunotherapy before the operation can make surgery easier, improve outcomes, and in some cases allow less aggressive procedures. Yet clinicians have long struggled with a fundamental uncertainty: after the treatment course ends, they cannot reliably know whether microscopic disease remains in the body. Imaging is too coarse to see residual cancer cells, and pathological examination of the resected specimen only becomes available after surgery, when the treatment window has already closed. This is precisely the gap that ctDNA promises to fill.

Circulating tumor DNA consists of small fragments of genetic material shed by cancer cells into the bloodstream as they die or actively release DNA. These fragments carry the same mutations that define the original tumor, which means that highly sensitive sequencing or digital PCR assays can distinguish them from the vast background of normal cell-free DNA released by healthy tissues. When a tumor responds completely to treatment, ctDNA levels typically fall below the limit of detection. When resistant clones survive, they continue to shed DNA, and a blood test can pick up that molecular signal long before any relapse becomes visible on a scan. The concept, often called a liquid biopsy, has already transformed follow-up care in colorectal cancer, and the new meta-analysis suggests its power extends across the entire spectrum of solid tumors treated before surgery.

The research team, led by Tao Jin and Ershuo Du with Kun Yang as corresponding author, systematically searched PubMed, Embase, and the Cochrane Library from database inception through October 2024. Two reviewers independently screened studies, extracted data, and graded methodological quality using the Newcastle-Ottawa Scale for cohort studies, the Cochrane Risk of Bias 2.0 tool for randomized trials, and ROBINS-I for non-randomized interventional studies. They used Stata 17.0 to pool hazard ratios for four survival endpoints: disease-free or relapse-free survival, overall survival, progression-free survival, and event-free survival. Publication bias was assessed with Begg’s test and funnel plots, and sensitivity analyses confirmed the robustness of the pooled estimates.

The headline numbers are striking. Patients with detectable ctDNA after neoadjuvant therapy had a hazard ratio of 8.60 for worse disease-free or relapse-free survival, meaning their risk of relapse or death was more than eight times higher than patients with negative tests. For overall survival the pooled hazard ratio was 7.00, for progression-free survival 3.85, and for event-free survival 3.40. In clinical terms, a positive ctDNA result after preoperative treatment behaves less like a modest risk modifier and more like a categorical dividing line between two very different disease trajectories. Few biomarkers in oncology, whether imaging-based, pathological, or molecular, achieve effect sizes of this magnitude.

The subgroup analyses reveal how consistently this signal holds across tumor types. In rectal cancer, positive ctDNA conferred a hazard ratio of 9.23 for disease-free or relapse-free survival and 9.72 for overall survival, a critical finding given the intense current interest in organ preservation strategies for rectal cancer, where clinicians weigh whether surgery can be safely omitted in patients who respond completely to chemoradiation. In urothelial cancer the association was even stronger, with hazard ratios of 27.92 for disease-free or relapse-free survival and 21.29 for overall survival. Esophageal cancer showed a disease-free survival hazard ratio of 10.83, while lung cancer patients with detectable ctDNA faced a 25-fold higher hazard of death. Breast cancer, pancreatic cancer, and colorectal liver metastases all showed significant associations as well, with hazard ratios ranging from roughly 3.6 to 11.7 depending on the endpoint.

The picture is not uniformly clear-cut, however, and the authors are careful to note that prognostic performance varies by tumor type and clinical context. In melanoma, the association between ctDNA detection and disease-free or relapse-free survival did not reach statistical significance, with a confidence interval crossing unity. In gastric cancer, ctDNA positivity predicted worse event-free survival with a hazard ratio of 3.00, but the overall survival association fell just short of significance. Ovarian cancer likewise showed a non-significant trend for overall survival. These exceptions likely reflect tumor biology, since melanoma and ovarian cancer shed DNA differently and often present with lower or more heterogeneous ctDNA fractions, as well as the smaller number of studies available for these subgroups.

Perhaps the most clinically important finding is that the signal survived every stratification the team applied. Whether studies were prospective or retrospective, whether blood was drawn before surgery or during follow-up, whether patients received immunotherapy or conventional regimens, whether ctDNA was measured by next-generation sequencing or droplet digital PCR, and whether the reported statistics came from univariate or multivariate models, positive ctDNA status consistently predicted worse outcomes on all four survival endpoints. This robustness matters because multivariate adjustment, in particular, demonstrates that ctDNA adds prognostic information beyond established clinical factors such as stage, grade, and pathological response. The consistency across detection platforms also suggests the finding is not an artifact of any single laboratory technology.

The implications for trial design and clinical practice are considerable. In rectal cancer, several ongoing trials are already using ctDNA clearance to select patients for a watch-and-wait approach instead of radical surgery, and this meta-analysis provides the strongest quantitative backing yet for that strategy. In breast and esophageal cancer, a positive post-treatment ctDNA result could identify patients who need escalation, such as additional chemotherapy, extended immunotherapy, or more aggressive surgery, while a negative result might safely allow de-escalation, sparing patients the toxicities of overtreatment. Because the test requires only a blood sample, it can be repeated at multiple time points, opening the door to real-time monitoring of treatment response rather than a single static assessment.

Challenges remain before ctDNA-guided decisions become standard of care. The included studies varied in assay sensitivity, timing of blood draws, and mutation panels, and most evidence is observational, so prospective randomized trials are still needed to prove that acting on ctDNA results actually improves survival rather than merely predicting it. The authors also note that the exceptional hazard ratios seen in some tumor types come from smaller subgroups with wide confidence intervals, and that heterogeneity across studies warrants cautious interpretation. Nevertheless, with 41 studies and more than 3,000 patients converging on the same conclusion, the evidence now strongly supports ctDNA as a prognostic biomarker for patients undergoing neoadjuvant therapy. What was once an experimental research tool is moving decisively toward the clinic, where a routine blood test could help decide, within days, whether the treatment a patient just endured has truly eradicated their cancer or whether invisible disease still lurks and demands a different plan.

Subject of Research: Prognostic value of circulating tumor DNA in solid tumor patients receiving neoadjuvant therapy

Article Title: Association between circulating tumor DNA and the prognosis of solid tumor patients receiving neoadjuvant therapy: a meta-analysis

Article References: Jin, T., Du, E., Fu, X., Cao, Y., Liang, P., Gou, Y., & Yang, K. (2026). Association between circulating tumor DNA and the prognosis of solid tumor patients receiving neoadjuvant therapy: a meta-analysis. BMC Cancer. https://doi.org/10.1186/s12885-026-16983-3

Image Credits: AI Generated

DOI: 10.1186/s12885-026-16983-3

Keywords: circulating tumor DNA, liquid biopsy, neoadjuvant therapy, meta-analysis, prognosis, solid tumors, biomarker, disease-free survival, overall survival, rectal cancer, breast cancer, precision oncology