The mean unidirectional equilibrium cellular water efflux rate constant, k_io, is obtained from shutter-speed [SS] [Dynamic‑Contrast-Enhanced] DCE‑MRI.¹ Cell suspension, perfused tissue, and animal model evidence indicates k_io differences report vital cell membrane ion pump Na⁺,K⁺-ATPase [NKA] turnover changes.¹ Human validation is desired. Correlating breast DCE-MRI biomarker maps with histopathology is particularly difficult, due to co-registration uncertainties. Here, we determine whole tumor averaged k_io values for 29 invasive ductal carcinomas [IDCs] obtained after 16-18 weeks of standard-of-care neoadjuvant chemotherapy [NACT]. Since this protocol includes post‑NACT surgery [lumpectomy, mostly mastectomy] just a few days later, it presents a unique opportunity to correlate MRI with whole tumor residual cancer burden [RCB; from pathology], avoiding co-registration problems. Twenty-eight consecutive grade 2-3 IDC subjects (in a protocol described previously¹) consented to 3T DCE-MRI studies before, during, and just after the NACT course [one had two tumors]. The bi-lateral, fat-suppressed 3D DCE‑MRI acquisitions included 96-128 slices, and temporal resolution 14.6-20.2 s.¹ The nominal voxel dimensions were (0.94‑1.1 mm)² x 1.4 mm. The data were analyzed with an SS model assuming a single ¹H₂O signal,^1,2 to yield the mean intracellular water lifetime, tau_i. For each subject, we averaged tau_i over all tumor voxels in all slices showing the lesion. We calculated <k_io>_tum as 1/<tau_i>_tum. [For a non‑normal tau_i distribution, this is not strictly true:¹ but the effect is small.] Figure 1 shows an axial T₁-weighted DCE-MR image of a grade 3 IDC patient before NACT. There is a clearly enhancing right breast tumor just anterior to an implant. Figure 2 shows the right mammary gland k_io map. Lesion k_io values are elevated compared with most of the normal-appearing gland – particularly in the tumor core. [The serpentine high k_io gland regions correlate with hypointense Fig. 1 regions: fat-suppressed k_io is artifactually increased in adipose-rich loci.] After NACT completion, histopathology was used to determine the in breast whole tumor RCB as 1.4[f_inv*(d₁d₂)^1/2]^0.17, where f_inv is the whole tumor volume fraction of invasive cells and (d₁d₂)^1/2 the geometric mean of orthogonal tumor diameters.³ Figure 3 plots the post-NACT, in vivo <k_io>_tum vs. ex vivo RCB, from just a few days later, for all tumors. [Those with RCB = 0 were declared complete responders by pathology (pCR).] There is a strong supra-exponential positive correlation: k_io increases with RCB, but the dependence is not intuitive. Why should k_io be curving upward as RCB increases? The clinical RCB equation has an intentional extensive nature: for a given f_inv, RCB increases with tumor size. However, k_io is an intensive parameter; independent of tumor size.¹ The ex vivo f_inv factor is the pathology quantity comparable to the post-NACT in vivo <k_io>_tum value. Figure 4 plots <k_io>_tum vs. f_inv [given as a volume %]. The supra‑hyperbolic curve shape now seen is sensible. The quantity <k_io>_tum increases with f_inv and begins to saturate as f_inv approaches 100%. The k_io precision is reasonable: for the four tumors with 0% f_inv [pCR subjects], <<k_io>_tum>₄ = 0.73 (± 0.20) s^-1. It is hard to know the <k_io>_tum value for 100% f_inv. Besides k_io uncertainty, the three largest f_inv values are reported as very rounded; 30%, 50%, and 75%. It does seem clear that, even after therapy, the 100% f_inv <k_io>_tum value is at least five-fold greater than that at 0% f_inv. Cancer cells exhibit overexpression of ion channels and reduced membrane potential.⁴ Thus, it is sensible this increases NKA turnover (in a futile re‑polarization attempt) and elevates k_io. The correlation seen in Fig. 4 is consistent with this. For one subject, a needle biopsy core obtained ~ 4 weeks pre-NACT was examined.¹ A region in the core micrograph exhibiting a high density of invasive carcinoma and lymphocyte cells was manually co‑registered with a tumor focus in the pre-NACT MRI having k_io ≈ 1.7 s^-1. The k_io biomarker shows promise for the metabolic assessment of in vivo breast and other tumors. It shows that lonidamine co‑therapy affects mitochondrial function.⁵ The k_io and v_i [intracellular volume fraction] biomarkers combine to distinguish cytotoxic from non-cytotoxic (but effective) targeted therapies.⁶