IHC Studies in Lymphoma

ls /home/daytona/workspace/ && python3 -c "import reportlab; print('reportlab ok…

mkdir -p /home/daytona/workspace/lymphoma-ihc

~/lymphoma-ihc/generate_pdf.py

from reportlab.lib.pagesizes import A4
from reportlab.lib import colors
from reportlab.lib.units import cm, mm
from reportlab.platypus import (
    SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle,
    HRFlowable, KeepTogether, PageBreak
)
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.enums import TA_CENTER, TA_LEFT
from reportlab.platypus import BaseDocTemplate, Frame, PageTemplate
from reportlab.lib.colors import HexColor

OUTPUT = "/home/daytona/workspace/lymphoma-ihc/Lymphoma_IHC_Quick_Reference.pdf"

# ── Colour palette ──────────────────────────────────────────────────────────
NAVY      = HexColor("#1B3A6B")
TEAL      = HexColor("#1A7A8A")
AMBER     = HexColor("#E67E22")
LIGHT_BG  = HexColor("#F0F4F8")
HEADER_BG = HexColor("#1B3A6B")
ROW_ALT   = HexColor("#E8F0F7")
WHITE     = colors.white
BLACK     = colors.black
POS_GREEN = HexColor("#1E7E34")
NEG_RED   = HexColor("#C0392B")
VAR_AMBER = HexColor("#D4860B")

# ── Styles ───────────────────────────────────────────────────────────────────
styles = getSampleStyleSheet()

TITLE_STYLE = ParagraphStyle("Title", fontName="Helvetica-Bold", fontSize=22,
                              textColor=WHITE, alignment=TA_CENTER, spaceAfter=4)
SUBTITLE_STYLE = ParagraphStyle("Subtitle", fontName="Helvetica", fontSize=11,
                                 textColor=HexColor("#B8D4E8"), alignment=TA_CENTER, spaceAfter=2)
DATE_STYLE = ParagraphStyle("Date", fontName="Helvetica-Oblique", fontSize=9,
                              textColor=HexColor("#B8D4E8"), alignment=TA_CENTER)

SECTION_STYLE = ParagraphStyle("Section", fontName="Helvetica-Bold", fontSize=13,
                                textColor=WHITE, alignment=TA_LEFT,
                                leftIndent=6, spaceAfter=0, spaceBefore=0)

BODY_STYLE = ParagraphStyle("Body", fontName="Helvetica", fontSize=8.5,
                              textColor=BLACK, spaceAfter=3, leading=12)
BODY_BOLD  = ParagraphStyle("BodyBold", fontName="Helvetica-Bold", fontSize=8.5,
                              textColor=BLACK, spaceAfter=3, leading=12)
CAPTION    = ParagraphStyle("Caption", fontName="Helvetica-Oblique", fontSize=7.5,
                              textColor=HexColor("#555555"), spaceAfter=4)
FOOTNOTE   = ParagraphStyle("Footnote", fontName="Helvetica-Oblique", fontSize=7,
                              textColor=HexColor("#666666"), spaceAfter=2)

# cell styles
TH_STYLE = ParagraphStyle("TH", fontName="Helvetica-Bold", fontSize=8,
                            textColor=WHITE, alignment=TA_CENTER, leading=10)
TD_STYLE = ParagraphStyle("TD", fontName="Helvetica", fontSize=7.8,
                            textColor=BLACK, alignment=TA_CENTER, leading=10)
TD_LEFT  = ParagraphStyle("TDL", fontName="Helvetica", fontSize=7.8,
                            textColor=BLACK, alignment=TA_LEFT, leading=10)
TD_BOLD  = ParagraphStyle("TDB", fontName="Helvetica-Bold", fontSize=7.8,
                            textColor=BLACK, alignment=TA_LEFT, leading=10)

def pos(text="+"):  return Paragraph(f'<font color="#1E7E34"><b>{text}</b></font>', TD_STYLE)
def neg(text="−"):  return Paragraph(f'<font color="#C0392B"><b>{text}</b></font>', TD_STYLE)
def var(text="+/−"): return Paragraph(f'<font color="#D4860B"><b>{text}</b></font>', TD_STYLE)
def th(text):        return Paragraph(f"<b>{text}</b>", TH_STYLE)
def td(text):        return Paragraph(text, TD_STYLE)
def tdl(text):       return Paragraph(text, TD_LEFT)
def tdb(text):       return Paragraph(f"<b>{text}</b>", TD_BOLD)

def section_bar(title):
    """Returns a coloured section header bar."""
    data = [[Paragraph(f"<b>{title}</b>", SECTION_STYLE)]]
    t = Table(data, colWidths=[17.7*cm])
    t.setStyle(TableStyle([
        ("BACKGROUND", (0,0), (-1,-1), TEAL),
        ("TOPPADDING",    (0,0), (-1,-1), 5),
        ("BOTTOMPADDING", (0,0), (-1,-1), 5),
        ("LEFTPADDING",   (0,0), (-1,-1), 8),
        ("ROUNDEDCORNERS", [3]),
    ]))
    return t

def base_table_style(header_rows=1):
    return [
        ("BACKGROUND",    (0,0),  (-1, header_rows-1), NAVY),
        ("TEXTCOLOR",     (0,0),  (-1, header_rows-1), WHITE),
        ("FONTNAME",      (0,0),  (-1, header_rows-1), "Helvetica-Bold"),
        ("FONTSIZE",      (0,0),  (-1,-1), 8),
        ("ROWBACKGROUNDS",(0, header_rows), (-1,-1), [WHITE, ROW_ALT]),
        ("GRID",          (0,0),  (-1,-1), 0.4, HexColor("#AAAAAA")),
        ("TOPPADDING",    (0,0),  (-1,-1), 4),
        ("BOTTOMPADDING", (0,0),  (-1,-1), 4),
        ("LEFTPADDING",   (0,0),  (-1,-1), 4),
        ("RIGHTPADDING",  (0,0),  (-1,-1), 4),
        ("VALIGN",        (0,0),  (-1,-1), "MIDDLE"),
    ]

# ── Header / footer callbacks ────────────────────────────────────────────────
def on_first_page(canvas, doc):
    canvas.saveState()
    w, h = A4
    # navy banner
    canvas.setFillColor(NAVY)
    canvas.rect(0, h-3.8*cm, w, 3.8*cm, fill=1, stroke=0)
    # teal accent stripe
    canvas.setFillColor(TEAL)
    canvas.rect(0, h-4.1*cm, w, 0.3*cm, fill=1, stroke=0)
    canvas.restoreState()

def on_later_pages(canvas, doc):
    canvas.saveState()
    w, h = A4
    # slim top bar
    canvas.setFillColor(NAVY)
    canvas.rect(0, h-1.2*cm, w, 1.2*cm, fill=1, stroke=0)
    canvas.setFont("Helvetica-Bold", 8)
    canvas.setFillColor(WHITE)
    canvas.drawString(1.5*cm, h-0.85*cm, "LYMPHOMA IHC — QUICK REFERENCE")
    canvas.drawRightString(w-1.5*cm, h-0.85*cm, f"Page {doc.page}")
    # footer
    canvas.setFillColor(NAVY)
    canvas.rect(0, 0, w, 0.8*cm, fill=1, stroke=0)
    canvas.setFont("Helvetica-Oblique", 7)
    canvas.setFillColor(HexColor("#B8D4E8"))
    canvas.drawCentredString(w/2, 0.28*cm,
        "Sources: Robbins Pathology · Harrison's Internal Medicine 22E · Quick Compendium of Clinical Pathology 5e · Henry's Lab Methods")
    canvas.restoreState()

# ── Build document ────────────────────────────────────────────────────────────
doc = SimpleDocTemplate(
    OUTPUT,
    pagesize=A4,
    leftMargin=1.5*cm, rightMargin=1.5*cm,
    topMargin=4.5*cm, bottomMargin=1.5*cm,
    title="Lymphoma IHC Quick Reference",
    author="Orris Medical AI",
)

story = []

# ── Cover header content ──────────────────────────────────────────────────────
story.append(Spacer(1, 0.2*cm))
story.append(Paragraph("LYMPHOMA IHC", TITLE_STYLE))
story.append(Paragraph("Quick Reference Guide for Immunohistochemical Stain Interpretation", SUBTITLE_STYLE))
story.append(Paragraph("Based on WHO Classification · Robbins Pathology · Harrison's 22E · Quick Compendium 5e", DATE_STYLE))
story.append(Spacer(1, 0.6*cm))

# ── LEGEND ────────────────────────────────────────────────────────────────────
legend_data = [
    [th("Symbol"), th("Meaning")],
    [pos("+"),  tdl("Positive / Expressed")],
    [neg("−"),  tdl("Negative / Not expressed")],
    [var("+/−"),tdl("Variable / Inconsistent expression")],
    [pos("dim+"),tdl("Weakly / dimly positive")],
    [td("~100%"),tdl("Near-100% proliferative index (Ki-67)")],
]
legend_table = Table(legend_data, colWidths=[2.5*cm, 6*cm])
legend_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("LEGEND"),
    Spacer(1, 0.15*cm),
    legend_table,
    Spacer(1, 0.5*cm),
]))

# ── SECTION 1: CORE LINEAGE PANEL ────────────────────────────────────────────
lineage_data = [
    [th("Marker"), th("Lineage / Role"), th("Key Diagnostic Use")],
    [tdb("CD45 (LCA)"), tdl("Pan-leukocyte"), tdl("ALL lymphomas +; Negative in classic Reed-Sternberg cells (HL)")],
    [tdb("CD20"),       tdl("B-cell"),       tdl("Mature B-cell lymphomas; rituximab therapeutic target")],
    [tdb("CD79a"),      tdl("B-cell"),       tdl("Sensitive B-cell marker; positive when CD20 lost (e.g. post-rituximab)")],
    [tdb("PAX5"),       tdl("B-cell TF"),    tdl("Weak + in Reed-Sternberg cells; confirms B origin in classic HL")],
    [tdb("CD3"),        tdl("Pan-T-cell"),   tdl("T-cell lymphomas; surface loss suggests neoplasia")],
    [tdb("CD5"),        tdl("T-cell + B subset"), tdl("CLL/SLL and MCL among B-cell lymphomas; absent in FL/DLBCL")],
    [tdb("CD10"),       tdl("Follicle center"), tdl("FL, DLBCL-GCB, Burkitt, AITL; absent in MCL/MZL/CLL")],
    [tdb("BCL6"),       tdl("GC B-cell TF"), tdl("FL, DLBCL-GCB; AITL (T-cells); NLPHL L&H cells")],
    [tdb("MUM1/IRF4"),  tdl("Post-GC/plasma cell"), tdl("DLBCL-ABC subtype; plasma cell myeloma; ALCL")],
    [tdb("BCL2"),       tdl("Anti-apoptotic"), tdl("FL (t14;18), DLBCL; Negative in Burkitt (key distinction)")],
    [tdb("Cyclin D1"),  tdl("Cell cycle"),   tdl("Hallmark of MCL t(11;14); also hairy cell leukemia")],
    [tdb("CD30"),       tdl("Activation marker"), tdl("Classic HL RS cells, ALCL, lymphomatoid papulosis")],
    [tdb("CD15"),       tdl("Granulocyte/RS"), tdl("Classic HL RS cells 75–85%; absent in NLPHL")],
    [tdb("Ki-67"),      tdl("Proliferation"), tdl("~100% Burkitt; high in DLBCL/ALCL; low in FL/SLL/MCL")],
    [tdb("EBV (LMP-1)"),tdl("Viral marker"), tdl("HL mixed-cellularity 70%, lymphocyte-depleted >90%; NK/T lymphoma")],
]
lineage_table = Table(lineage_data, colWidths=[2.6*cm, 3.1*cm, 12*cm])
lineage_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("1 · CORE LINEAGE & DIAGNOSTIC MARKERS"),
    Spacer(1, 0.15*cm),
    lineage_table,
    Spacer(1, 0.5*cm),
]))

# ── SECTION 2: MASTER IHC TABLE ──────────────────────────────────────────────
# Columns: Lymphoma | CD20 | CD5 | CD10 | CD23 | CycD1 | BCL2 | BCL6 | CD30 | CD15 | CD45 | Ki-67
master_headers = [th(x) for x in [
    "Lymphoma", "CD20","CD5","CD10","CD23","CyclinD1","BCL2","BCL6","CD30","CD15","CD45","Ki-67"
]]
master_data = [master_headers,
    [tdb("SLL / CLL"),    pos("dim+"), pos(), neg(),  pos(),  neg(),   pos(),   neg(),  neg(),  neg(),  pos(),  td("Low")],
    [tdb("MCL"),          pos(),       pos(), neg(),  neg(),  pos("✓"),pos(),   neg(),  neg(),  neg(),  pos(),  td("Var")],
    [tdb("Follicular"),   pos(),       neg(), pos(),  neg(),  neg(),   pos("✓"),pos(),  neg(),  neg(),  pos(),  td("Low")],
    [tdb("MALT/MZL"),     pos(),       neg(), neg(),  neg(),  neg(),   var(),   neg(),  neg(),  neg(),  pos(),  td("Low")],
    [tdb("DLBCL-GCB"),    pos(),       neg(), pos(),  neg(),  neg(),   var(),   pos(),  neg(),  neg(),  pos(),  td("High")],
    [tdb("DLBCL-ABC"),    pos(),       neg(), neg(),  neg(),  neg(),   var(),   neg(),  neg(),  neg(),  pos(),  td("High")],
    [tdb("Burkitt"),      pos(),       neg(), pos(),  neg(),  neg(),   neg("✓"),pos(),  neg(),  neg(),  pos(),  td("~100%")],
    [tdb("Classic HL"),   neg(),       neg(), neg(),  neg(),  neg(),   neg(),   neg(),  pos("✓"),pos("✓"),neg("✓"),td("Var")],
    [tdb("NLPHL"),        pos("✓"),    neg(), neg(),  neg(),  neg(),   neg(),   pos(),  neg(),  neg(),  pos(),  td("Low")],
    [tdb("ALCL (ALK+)"),  neg(),       neg(), neg(),  neg(),  neg(),   neg(),   neg(),  pos("✓"),neg(),  pos(),  td("High")],
    [tdb("PTCL-NOS"),     neg(),       neg(), neg(),  neg(),  neg(),   neg(),   neg(),  neg(),  neg(),  pos(),  td("High")],
    [tdb("NK/T-cell"),    neg(),       neg(), neg(),  neg(),  neg(),   neg(),   neg(),  neg(),  neg(),  pos(),  td("High")],
]

# Highlight special "key" cells
col_widths = [2.8*cm, 1.3*cm, 1.1*cm, 1.1*cm, 1.15*cm, 1.5*cm, 1.15*cm, 1.15*cm, 1.15*cm, 1.15*cm, 1.15*cm, 1.3*cm]
master_table = Table(master_data, colWidths=col_widths)
ts = base_table_style()
# mark key diagnostics with light gold
key_cells = [
    (5,2),(5,3),(5,4),(5,5),   # MCL cyclin D1
    (6,3),(6,7),                # FL CD10, BCL2
    (9,11),                     # Burkitt Ki67
    (8,9),(8,10),(8,11),        # Classic HL CD30 CD15 CD45
    (9,8),                      # NLPHL CD20
    (10,9),                     # ALCL CD30
]
for (row,col) in key_cells:
    ts.append(("BACKGROUND", (col,row), (col,row), HexColor("#FFF3CD")))

master_table.setStyle(TableStyle(ts))

story.append(KeepTogether([
    section_bar("2 · MASTER IHC PROFILE TABLE"),
    Spacer(1, 0.15*cm),
    master_table,
    Spacer(1, 0.15*cm),
    Paragraph("✓ = hallmark/diagnostic marker  |  Highlighted cells = key distinguishing markers", CAPTION),
    Spacer(1, 0.4*cm),
]))

# ── SECTION 3: HODGKIN LYMPHOMA ───────────────────────────────────────────────
hl_data = [
    [th("Subtype"), th("% of HL"), th("RS Cell Variant"), th("IHC Profile"), th("EBV"), th("Notes")],
    [tdb("Nodular Sclerosis"),     td("65–70%"), tdl("Lacunar cells"),           tdl("CD30+, CD15+, PAX5+w, CD45−"), td("Rare"),   tdl("Young adults; mediastinal disease")],
    [tdb("Mixed Cellularity"),     td("20–25%"), tdl("Classic RS cells"),        tdl("CD30+, CD15+, PAX5+w, CD45−"), td("~70%"),   tdl("Older males; advanced stage")],
    [tdb("Lymphocyte-Rich"),       td("~5%"),    tdl("Mononuclear variants"),    tdl("CD30+, CD15+, PAX5+w, CD45−"), td("~40%"),   tdl("Good prognosis")],
    [tdb("Lymphocyte-Depleted"),   td("<5%"),    tdl("Pleomorphic RS"),          tdl("CD30+, CD15+, PAX5+w, CD45−"), td(">90%"),   tdl("Poorest prognosis; rule out ALCL")],
    [tdb("NLPHL (non-classic)"),   td("~5%"),    tdl("Popcorn/L&H cells"),       tdl("CD20+, BCL6+, CD30−, CD15−, CD45+"), td("Rare"), tdl("Different entity; better prognosis")],
]
hl_table = Table(hl_data, colWidths=[3.4*cm, 1.6*cm, 2.8*cm, 4.8*cm, 1.2*cm, 3.9*cm])
hl_ts = base_table_style()
# NLPHL row highlight in amber background
for col in range(6):
    hl_ts.append(("BACKGROUND", (col, 5), (col, 5), HexColor("#FFF3CD")))
hl_table.setStyle(TableStyle(hl_ts))

story.append(KeepTogether([
    section_bar("3 · HODGKIN LYMPHOMA — SUBTYPE IHC"),
    Spacer(1, 0.15*cm),
    hl_table,
    Spacer(1, 0.15*cm),
    Paragraph("w = weakly positive  |  NLPHL = Nodular Lymphocyte Predominant HL (highlighted — distinct from classic HL)", CAPTION),
    Spacer(1, 0.4*cm),
]))

# ── SECTION 4: DLBCL SUBTYPING — HANS ALGORITHM ─────────────────────────────
hans_data = [
    [th("Step"), th("Marker"), th("Threshold"), th("Result")],
    [td("1"), tdb("CD10"), tdl("> 30% positive"), tdl("→ GCB type (stop)")],
    [td("2"), tdb("BCL6"), tdl("Positive"), tdl("→ Proceed to step 3; if negative → non-GCB (ABC)")],
    [td("3"), tdb("MUM1"), tdl("Positive"), tdl("→ non-GCB (ABC); if negative → GCB")],
]
hans_table = Table(hans_data, colWidths=[1.5*cm, 2.5*cm, 4.5*cm, 9.2*cm])
hans_table.setStyle(TableStyle(base_table_style()))

combo_data = [
    [th("CD10"), th("BCL6"), th("MUM1"), th("Phenotype"), th("Common Genetics"), th("Prognosis")],
    [pos(), pos(), neg(), tdb("GCB"),       tdl("t(14;18) BCL2; BCL6 rare"), tdl("Better (R-CHOP sensitive)")],
    [pos(), neg(), neg(), tdb("GCB"),       tdl("t(14;18) BCL2"),            tdl("Better")],
    [neg(), pos(), neg(), tdb("GCB"),       tdl("BCL6 rearrangement"),       tdl("Better")],
    [neg(), pos(), pos(), tdb("non-GCB (ABC)"), tdl("BCL6 t(3;X)"),         tdl("Worse")],
    [neg(), neg(), pos(), tdb("non-GCB (ABC)"), tdl("NF-κB activation"),    tdl("Worse")],
]
combo_table = Table(combo_data, colWidths=[1.8*cm, 1.8*cm, 1.8*cm, 3.2*cm, 5.5*cm, 3.6*cm])
combo_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("4 · DLBCL SUBTYPING — HANS ALGORITHM (IHC)"),
    Spacer(1, 0.15*cm),
    Paragraph("<b>Step-wise evaluation:</b>", BODY_BOLD),
    Spacer(1, 0.1*cm),
    hans_table,
    Spacer(1, 0.25*cm),
    Paragraph("<b>Full combination matrix:</b>", BODY_BOLD),
    Spacer(1, 0.1*cm),
    combo_table,
    Spacer(1, 0.15*cm),
    Paragraph(
        "Double-hit lymphoma: MYC rearrangement + BCL2 and/or BCL6 rearrangement — confirm by FISH. "
        "Extremely poor prognosis (median OS 12–18 months). MYC protein >40% + BCL2 >50% by IHC = "
        '"double-expressor" — also adverse even without rearrangements.',
        BODY_STYLE),
    Spacer(1, 0.4*cm),
]))

# ── SECTION 5: T-CELL / NK-CELL LYMPHOMAS ────────────────────────────────────
tcell_data = [
    [th("Lymphoma"), th("Key Positive Markers"), th("Key Negative"), th("Special Marker"), th("Notes")],
    [tdb("PTCL-NOS"),    tdl("CD3, CD4 or CD8"), tdl("No single defining marker"), tdl("Aberrant T-cell Ag loss"), tdl("Diagnosis of exclusion")],
    [tdb("ALCL ALK+"),   tdl("CD30 (strong), EMA"), tdl("CD3 often −"), tdl("ALK (nuclear+cytoplasmic)"), tdl("Better prognosis; t(2;5) NPM-ALK")],
    [tdb("ALCL ALK−"),   tdl("CD30 (strong), EMA"), tdl("ALK−"),        tdl("DUSP22 or TP63 by FISH"), tdl("Worse prognosis")],
    [tdb("AITL"),        tdl("CD3, CD4, PD-1, CXCL13, CD10"), tdl("CD8"), tdl("CD21+ expanded FDC meshwork, EBER+"), tdl("TFH cell origin")],
    [tdb("NK/T-cell"),   tdl("CD56, CD2, cytoCD3, Granzyme B"), tdl("sCD3−, CD5−"), tdl("EBV (EBER) in situ hybridisation +"), tdl("Nasal type; aggressive")],
    [tdb("Hepatosplenic"), tdl("CD3, CD56"), tdl("CD4−, CD8−"), tdl("TCRγδ+"), tdl("Young males; sinusoidal")],
    [tdb("ATLL"),        tdl("CD3, CD4, CD25"), tdl("CD7 (often lost)"), tdl("HTLV-1 serology"), tdl("FoxP3+ (regulatory T-cell)")],
    [tdb("MF / Sézary"), tdl("CD3, CD4"), tdl("CD7, CD26 loss"), tdl("CD4:CD8 ratio >10 in blood"), tdl("Epidermotropic; indolent")],
]
tcell_table = Table(tcell_data, colWidths=[2.8*cm, 4.2*cm, 2.5*cm, 4.4*cm, 3.8*cm])
tcell_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("5 · T-CELL & NK-CELL LYMPHOMAS"),
    Spacer(1, 0.15*cm),
    tcell_table,
    Spacer(1, 0.15*cm),
    Paragraph("AITL = Angioimmunoblastic T-cell Lymphoma  |  ALCL = Anaplastic Large Cell Lymphoma  |  MF = Mycosis Fungoides  |  TFH = T follicular helper", CAPTION),
    Spacer(1, 0.4*cm),
]))

# ── PAGE 2: DIFFERENTIAL DIAGNOSIS PANELS ────────────────────────────────────
story.append(PageBreak())

# ── SECTION 6: KEY DIFFERENTIALS ──────────────────────────────────────────────
diff_data = [
    [th("Differential"), th("CD5"), th("CD10"), th("CD23"), th("Cyclin D1"), th("BCL2"), th("SOX11"), th("Key Distinguisher")],
    [tdb("CLL/SLL"),   pos("dim"), neg(), pos(), neg(), pos(), neg(), tdl("CD23+, CD5+, dim CD20, dim sIg")],
    [tdb("MCL"),       pos(),      neg(), neg(), pos("✓"),pos(),pos("✓"),tdl("Cyclin D1+ (FISH if negative); SOX11+")],
    [tdb("FL"),        neg(),      pos(), neg(), neg(), pos("✓"),neg(),tdl("CD10+, BCL2+, follicular pattern, t(14;18)")],
    [tdb("MZL/MALT"),  neg(),      neg(), neg(), neg(), var(), neg(), tdl("All negative — diagnosis by exclusion + morphology")],
]
diff_table = Table(diff_data, colWidths=[2.6*cm, 1.4*cm, 1.4*cm, 1.5*cm, 2.2*cm, 1.4*cm, 1.5*cm, 6.2*cm])
diff_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("6 · DIFFERENTIAL: SMALL B-CELL LYMPHOMAS"),
    Spacer(1, 0.15*cm),
    diff_table,
    Spacer(1, 0.4*cm),
]))

# ── Burkitt vs DLBCL ──────────────────────────────────────────────────────────
bvd_data = [
    [th("Feature"),        th("Burkitt Lymphoma"),               th("DLBCL")],
    [tdb("CD20"),          pos(),                                pos()],
    [tdb("CD10"),          pos(),                                var()],
    [tdb("BCL6"),          pos(),                                var()],
    [tdb("BCL2"),          neg("✓ negative"),                    var()],
    [tdb("MUM1"),          neg(),                                var()],
    [tdb("Ki-67"),         td("~100% ✓"),                        td("60–90%")],
    [tdb("MYC"),           tdl("Rearranged (FISH) 100%"),        tdl("~10% rearranged")],
    [tdb("Morphology"),    tdl('"Starry sky" — tingible-body macrophages'), tdl("Large atypical cells, necrosis")],
    [tdb("Translocation"), tdl("t(8;14), t(2;8), t(8;22)"),     tdl("t(14;18) in GCB; BCL6 in ABC")],
]
bvd_table = Table(bvd_data, colWidths=[3.2*cm, 7.25*cm, 7.25*cm])
bvd_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("7 · DIFFERENTIAL: BURKITT vs DLBCL"),
    Spacer(1, 0.15*cm),
    bvd_table,
    Spacer(1, 0.4*cm),
]))

# ── Classic HL vs NLPHL ────────────────────────────────────────────────────────
hl_diff_data = [
    [th("Feature"),    th("Classic HL"),                     th("NLPHL")],
    [tdb("RS/Tumor cell"), tdl("Classic RS, lacunar, mononuclear"), tdl("Popcorn/L&H cells")],
    [tdb("CD20"),      neg(),                                pos("✓")],
    [tdb("CD45"),      neg("✓"),                             pos()],
    [tdb("CD15"),      pos("✓"),                             neg("✓")],
    [tdb("CD30"),      pos("✓"),                             neg("rare+")],
    [tdb("PAX5"),      pos("weak"),                          pos("strong")],
    [tdb("BCL6"),      neg(),                                pos("✓")],
    [tdb("EBV"),       var("+/− (subtype dependent)"),       neg("Rare")],
    [tdb("EMA"),       neg(),                                pos("+/−")],
    [tdb("Background"),tdl("T-cells, eosinophils, plasma cells"), tdl("B-cell-rich nodules, FDC meshwork")],
    [tdb("Prognosis"), tdl("Excellent with chemo-RT"),      tdl("Excellent; late relapses possible")],
]
hl_diff_table = Table(hl_diff_data, colWidths=[3.2*cm, 7.25*cm, 7.25*cm])
hl_diff_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("8 · DIFFERENTIAL: CLASSIC HL vs NLPHL"),
    Spacer(1, 0.15*cm),
    hl_diff_table,
    Spacer(1, 0.4*cm),
]))

# ── SECTION 9: PRACTICAL TIPS ─────────────────────────────────────────────────
tips = [
    ("<b>CD45 first:</b> A CD45-negative large cell tumour raises HL, germ cell tumour, carcinoma, or melanoma "
     "before NHL. Always confirm with CD30/CD15 (HL) or keratins/PLAP (carcinoma/GCT)."),
    ("<b>BCL2 in follicular lymphoma:</b> BCL2+ in neoplastic follicles vs. BCL2− in reactive follicles. "
     "Confirm t(14;18) by FISH when in doubt."),
    ("<b>Burkitt BCL2:</b> BCL2 negativity with Ki-67 ~100% virtually excludes DLBCL and is a red flag for Burkitt. "
     "Always confirm MYC rearrangement by FISH."),
    ("<b>Double-expressor DLBCL:</b> MYC >40% + BCL2 >50% by IHC without FISH-confirmed rearrangement still "
     "confers adverse prognosis but does not define 'double-hit lymphoma'. FISH is needed for the latter."),
    ("<b>ALK in ALCL:</b> ALK+ ALCL (t(2;5), NPM-ALK fusion) — nuclear+cytoplasmic staining. "
     "ALK− ALCL — identical morphology and CD30 positivity but ALK absent; worse prognosis."),
    ("<b>CD5+ B-cell lymphoma:</b> MCL (cyclin D1+, SOX11+) vs CLL/SLL (CD23+, dim sIg). "
     "Rarely, DLBCL can be CD5+ — aggressive behaviour."),
    ("<b>T-cell marker loss:</b> In T-cell lymphomas, aberrant loss of CD3, CD5, or CD7 supports neoplasia. "
     "Normal T-cells express all three."),
    ("<b>EBV in NK/T-cell lymphoma:</b> EBER (EBV-encoded RNA) by in situ hybridisation is the preferred method "
     "— more sensitive than LMP-1 by IHC."),
    ("<b>Hairy Cell Leukaemia:</b> CD20+, CD11c+, CD25+, CD103+ (TRAP+). CD5−, CD23−. "
     "BRAF V600E IHC (VE1 antibody) positive in >95%."),
]

tips_story = [
    section_bar("9 · PRACTICAL DIAGNOSTIC TIPS"),
    Spacer(1, 0.15*cm),
]
for i, tip in enumerate(tips, 1):
    tips_story.append(Paragraph(f"<b>{i}.</b>  {tip}", BODY_STYLE))

story.append(KeepTogether(tips_story))
story.append(Spacer(1, 0.5*cm))

# ── SECTION 10: QUICK ALGORITHM BOXES ─────────────────────────────────────────
algo_data = [
    [th("Clinical Scenario"), th("First IHC Panel"), th("Decision")],
    [tdl("Undifferentiated large cell"), tdl("CD45, CK, S100, CD30"), tdl("CD45+ → lymphoma panel | CD45− → carcinoma/melanoma/HL")],
    [tdl("Small B-cell lymphoma"), tdl("CD5, CD10, CD23, Cyclin D1"), tdl("See Section 6 algorithm above")],
    [tdl("Large B-cell lymphoma"), tdl("CD20, CD10, BCL6, MUM1, BCL2, Ki-67"), tdl("Apply Hans algorithm; add MYC IHC if high grade")],
    [tdl("Mediastinal mass"), tdl("CD20, CD30, CD15, CD45, PAX5"), tdl("HL vs PMBCL (CD20+ CD23+ BCL6+ MAL+) vs T-LBL")],
    [tdl("Suspect Burkitt"), tdl("CD20, CD10, BCL6, BCL2, Ki-67, MYC FISH"), tdl("BCL2−, Ki-67 ~100%, MYC FISH+ = Burkitt")],
    [tdl("Suspect T-cell NHL"), tdl("CD3, CD4, CD8, CD5, CD7, CD30, ALK"), tdl("Loss of CD5/CD7 supports neoplasia; CD30+ → ALCL panel")],
    [tdl("Skin lymphoma"), tdl("CD3, CD4, CD8, CD20, CD30, CD56"), tdl("CD4+ epidermotropic = MF; CD30+ = lymphomatoid papulosis/pcALCL")],
]
algo_table = Table(algo_data, colWidths=[4.0*cm, 5.5*cm, 8.2*cm])
algo_table.setStyle(TableStyle(base_table_style()))

story.append(KeepTogether([
    section_bar("10 · CLINICAL SCENARIO — PANEL SELECTION GUIDE"),
    Spacer(1, 0.15*cm),
    algo_table,
    Spacer(1, 0.4*cm),
]))

# ── FOOTER DISCLAIMER ─────────────────────────────────────────────────────────
story.append(HRFlowable(width="100%", thickness=0.5, color=NAVY))
story.append(Spacer(1, 0.15*cm))
story.append(Paragraph(
    "This quick-reference document is intended for educational and clinical support purposes. "
    "IHC interpretation must always be correlated with morphology, clinical context, and molecular findings. "
    "Sources: Robbins, Cotran & Kumar Pathologic Basis of Disease; Harrison's Principles of Internal Medicine 22E; "
    "Quick Compendium of Clinical Pathology 5th Edition; Henry's Clinical Diagnosis and Management by Laboratory Methods; "
    "Murray & Nadel's Textbook of Respiratory Medicine. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues.",
    FOOTNOTE))

# ── Build PDF ─────────────────────────────────────────────────────────────────
doc.build(story,
          onFirstPage=on_first_page,
          onLaterPages=on_later_pages)

print(f"PDF generated: {OUTPUT}")

python3 /home/daytona/workspace/lymphoma-ihc/generate_pdf.py 2>&1

ls -lh /home/daytona/workspace/lymphoma-ihc/Lymphoma_IHC_Quick_Reference.pdf